Convolutional computing using multilayered analysis engine (2024)

This application claims the benefit of U.S. provisional patent applications “Vehicle Interior Object Management” Ser. No. 62/893,298, filed Aug. 29, 2019, “Deep Learning In Situ Retraining” Ser. No. 62/925,990, filed Oct. 25, 2019, “Data Versioning for Neural Network Training” Ser. No. 62/926,009, filed Oct. 25, 2019, “Synthetic Data Augmentation for Neural Network Training” Ser. No. 62/954,819, filed Dec. 30, 2019, “Synthetic Data for Neural Network Training Using Vectors” Ser. No. 62/954,833, filed Dec. 30, 2019, and “Autonomous Vehicle Control Using Longitudinal Profile Generation” Ser. No. 62/955,493, filed Dec. 31, 2019.

This application is also a continuation-in-part of U.S. patent application “Vehicle Manipulation Using Convolutional Image Processing” Ser. No. 15/910,385, filed Mar. 2, 2018, which claims the benefit of U.S. provisional patent applications “Image Analysis for Two-sided Data Hub” Ser. No. 62/469,591, filed Mar. 10, 2017, “Vehicle Artificial Intelligence Evaluation of Mental States” Ser. No. 62/503,485, filed May 9, 2017, “Image Analysis for Emotional Metric Generation” Ser. No. 62/524,606, filed Jun. 25, 2017, “Image Analysis and Representation for Emotional Metric Threshold Evaluation” Ser. No. 62/541,847, filed Aug. 7, 2017, “Multimodal Machine Learning for Emotion Metrics” Ser. No. 62/557,460, filed Sep. 12, 2017, “Speech Analysis for Cross-Language Mental State Identification” Ser. No. 62/593,449, filed Dec. 1, 2017, “Avatar Image Animation Using Translation Vectors” Ser. No. 62/593,440, filed Dec. 1, 2017, “Directed Control Transfer for Autonomous Vehicles” Ser. No. 62/611,780, filed Dec. 29, 2017, and “Cognitive State Vehicle Navigation Based on Image Processing” Ser. No. 62/625,274, filed Feb. 1, 2018.

The U.S. patent application “Vehicle Manipulation Using Convolutional Image Processing” Ser. No. 15/910,385, filed Mar. 2, 2018 is a continuation-in-part of U.S. patent application “Computer Based Convolutional Processing for Image Analysis” Ser. No. 15/666,048, filed Aug. 1, 2017, which claims the benefit of U.S. provisional patent applications “Deep Convolutional Neural Network Analysis of Images for Mental States” Ser. No. 62/370,421, filed Aug. 3, 2016, “Image Analysis Framework using Remote Learning with Deployable Artifact” Ser. No. 62/439,928, filed Dec. 29, 2016, “Audio Analysis Learning using Video Data” Ser. No. 62/442,325, filed Jan. 4, 2017, “Vehicle Manipulation using Occupant Image Analysis” Ser. No. 62/448,448, filed Jan. 20, 2017, “Smart Toy Interaction using Image Analysis” Ser. No. 62/442,291, filed Jan. 4, 2017, “Image Analysis for Two-sided Data Hub” Ser. No. 62/469,591, filed Mar. 10, 2017, “Vehicle Artificial Intelligence Evaluation of Mental States” Ser. No. 62/503,485, filed May 9, 2017, and “Image Analysis for Emotional Metric Generation” Ser. No. 62/524,606, filed Jun. 25, 2017.

The patent application “Computer Based Convolutional Processing for Image Analysis” Ser. No. 15/666,048, filed Aug. 1, 2017 is also a continuation-in-part of U.S. patent application “Image Analysis using Sub-sectional Component Evaluation to Augment Classifier Usage” Ser. No. 15/395,750, filed Dec. 30, 2016, which claims the benefit of U.S. provisional patent applications “Image Analysis Using Sub-Sectional Component Evaluation to Augment Classifier Usage” Ser. No. 62/273,896, filed Dec. 31, 2015, “Analytics for Live Streaming Based on Image Analysis within a Shared Digital Environment” Ser. No. 62/301,558, filed Feb. 29, 2016, and “Deep Convolutional Neural Network Analysis of Images for Mental States” Ser. No. 62/370,421, filed Aug. 3, 2016.

The patent application “Image Analysis using Sub-sectional Component Evaluation to Augment Classifier Usage” Ser. No. 15/395,750, filed Dec. 30, 2016, is also a continuation-in-part of U.S. patent application “Mental State Event Signature Usage” Ser. No. 15/262,197, filed Sep. 12, 2016, which claims the benefit of U.S. provisional patent applications “Mental State Event Signature Usage” Ser. No. 62/217,872, filed Sep. 12, 2015, “Image Analysis In Support of Robotic Manipulation” Ser. No. 62/222,518, filed Sep. 23, 2015, “Analysis of Image Content with Associated Manipulation of Expression Presentation” Ser. No. 62/265,937, filed Dec. 10, 2015, “Image Analysis Using Sub-Sectional Component Evaluation To Augment Classifier Usage” Ser. No. 62/273,896, filed Dec. 31, 2015, “Analytics for Live Streaming Based on Image Analysis within a Shared Digital Environment” Ser. No. 62/301,558, filed Feb. 29, 2016, and “Deep Convolutional Neural Network Analysis of Images for Mental States” Ser. No. 62/370,421, filed Aug. 3, 2016.

The patent application “Mental State Event Signature Usage” Ser. No. 15/262,197, filed Sep. 12, 2016, is also a continuation-in-part of U.S. patent application “Mental State Event Definition Generation” Ser. No. 14/796,419, filed Jul. 10, 2015, which claims the benefit of U.S. provisional patent applications “Mental State Event Definition Generation” Ser. No. 62/023,800, filed Jul. 11, 2014, “Facial Tracking with Classifiers” Ser. No. 62/047,508, filed Sep. 8, 2014, “Semiconductor Based Mental State Analysis” Ser. No. 62/082,579, filed Nov. 20, 2014, and “Viewership Analysis Based On Facial Evaluation” Ser. No. 62/128,974, filed Mar. 5, 2015.

The patent application “Mental State Event Definition Generation” Ser. No. 14/796,419, filed Jul. 10, 2015 is also a continuation-in-part of U.S. patent application “Mental State Analysis Using Web Services” Ser. No. 13/153,745, filed Jun. 6, 2011, which claims the benefit of U.S. provisional patent applications “Mental State Analysis Through Web Based Indexing” Ser. No. 61/352,166, filed Jun. 7, 2010, “Measuring Affective Data for Web-Enabled Applications” Ser. No. 61/388,002, filed Sep. 30, 2010, “Sharing Affect Across a Social Network” Ser. No. 61/414,451, filed Nov. 17, 2010, “Using Affect Within a Gaming Context” Ser. No. 61/439,913, filed Feb. 6, 2011, “Recommendation and Visualization of Affect Responses to Videos” Ser. No. 61/447,089, filed Feb. 27, 2011, “Video Ranking Based on Affect” Ser. No. 61/447,464, filed Feb. 28, 2011, and “Baseline Face Analysis” Ser. No. 61/467,209, filed Mar. 24, 2011.

The patent application “Mental State Event Definition Generation” Ser. No. 14/796,419, filed Jul. 10, 2015 is also a continuation-in-part of U.S. patent application “Mental State Analysis Using an Application Programming Interface” Ser. No. 14/460,915, Aug. 15, 2014, which claims the benefit of U.S. provisional patent applications “Application Programming Interface for Mental State Analysis” Ser. No. 61/867,007, filed Aug. 16, 2013, “Mental State Analysis Using an Application Programming Interface” Ser. No. 61/924,252, filed Jan. 7, 2014, “Heart Rate Variability Evaluation for Mental State Analysis” Ser. No. 61/916,190, filed Dec. 14, 2013, “Mental State Analysis for Norm Generation” Ser. No. 61/927,481, filed Jan. 15, 2014, “Expression Analysis in Response to Mental State Express Request” Ser. No. 61/953,878, filed Mar. 16, 2014, “Background Analysis of Mental State Expressions” Ser. No. 61/972,314, filed Mar. 30, 2014, and “Mental State Event Definition Generation” Ser. No. 62/023,800, filed Jul. 11, 2014.

The patent application “Mental State Analysis Using an Application Programming Interface” Ser. No. 14/460,915, Aug. 15, 2014 is also a continuation-in-part of U.S. patent application “Mental State Analysis Using Web Services” Ser. No. 13/153,745, filed Jun. 6, 2011, which claims the benefit of U.S. provisional patent applications “Mental State Analysis Through Web Based Indexing” Ser. No. 61/352,166, filed Jun. 7, 2010, “Measuring Affective Data for Web-Enabled Applications” Ser. No. 61/388,002, filed Sep. 30, 2010, “Sharing Affect Across a Social Network” Ser. No. 61/414,451, filed Nov. 17, 2010, “Using Affect Within a Gaming Context” Ser. No. 61/439,913, filed Feb. 6, 2011, “Recommendation and Visualization of Affect Responses to Videos” Ser. No. 61/447,089, filed Feb. 27, 2011, “Video Ranking Based on Affect” Ser. No. 61/447,464, filed Feb. 28, 2011, and “Baseline Face Analysis” Ser. No. 61/467,209, filed Mar. 24, 2011.

The U.S. patent application “Vehicle Manipulation Using Convolutional Image Processing” Ser. No. 15/910,385, filed Mar. 2, 2018 is also a continuation-in-part of U.S. patent application “Vehicular Cognitive Data Collection Using Multiple Devices” Ser. No. 15/886,275, filed Feb. 1, 2018, which claims the benefit of U.S. provisional patent applications “Image Analysis for Two-sided Data Hub” Ser. No. 62/469,591, filed Mar. 10, 2017, “Vehicle Artificial Intelligence Evaluation of Mental States” Ser. No. 62/503,485, filed May 9, 2017, “Image Analysis for Emotional Metric Generation” Ser. No. 62/524,606, filed Jun. 25, 2017, “Image Analysis and Representation for Emotional Metric Threshold Evaluation” Ser. No. 62/541,847, filed Aug. 7, 2017, “Multimodal Machine Learning for Emotion Metrics” Ser. No. 62/557,460, filed Sep. 12, 2017, “Speech Analysis for Cross-Language Mental State Identification” Ser. No. 62/593,449, filed Dec. 1, 2017, “Avatar Image Animation Using Translation Vectors” Ser. No. 62/593,440, filed Dec. 1, 2017, and “Directed Control Transfer for Autonomous Vehicles” Ser. No. 62/611,780, filed Dec. 29, 2017.

The patent application “Vehicular Cognitive Data Collection Using Multiple Devices” Ser. No. 15/886,275, filed Feb. 1, 2018 is also a continuation-in-part of U.S. patent application “Collection of Affect Data from Multiple Mobile Devices” Ser. No. 14/144,413, filed Dec. 30, 2013, which claims the benefit of U.S. provisional patent applications “Optimizing Media Based on Mental State Analysis” Ser. No. 61/747,651, filed Dec. 31, 2012, “Collection of Affect Data from Multiple Mobile Devices” Ser. No. 61/747,810, filed Dec. 31, 2012, “Mental State Analysis Using Heart Rate Collection Based on Video Imagery” Ser. No. 61/793,761, filed Mar. 15, 2013, “Mental State Data Tagging for Data Collected from Multiple Sources” Ser. No. 61/790,461, filed Mar. 15, 2013, “Mental State Analysis Using Blink Rate” Ser. No. 61/789,038, filed Mar. 15, 2013, “Mental State Well Being Monitoring” Ser. No. 61/798,731, filed Mar. 15, 2013, and “Personal Emotional Profile Generation” Ser. No. 61/844,478, filed Jul. 10, 2013.

The patent application “Collection of Affect Data from Multiple Mobile Devices” Ser. No. 14/144,413, filed Dec. 30, 2013 is also a continuation-in-part of U.S. patent application “Mental State Analysis Using Web Services” Ser. No. 13/153,745, filed Jun. 6, 2011, which claims the benefit of U.S. provisional patent applications “Mental State Analysis Through Web Based Indexing” Ser. No. 61/352,166, filed Jun. 7, 2010, “Measuring Affective Data for Web-Enabled Applications” Ser. No. 61/388,002, filed Sep. 30, 2010, “Sharing Affect Data Across a Social Network” Ser. No. 61/414,451, filed Nov. 17, 2010, “Using Affect Within a Gaming Context” Ser. No. 61/439,913, filed Feb. 6, 2011, “Recommendation and Visualization of Affect Responses to Videos” Ser. No. 61/447,089, filed Feb. 27, 2011, “Video Ranking Based on Affect” Ser. No. 61/447,464, filed Feb. 28, 2011, and “Baseline Face Analysis” Ser. No. 61/467,209, filed Mar. 24, 2011.

The patent application “Collection of Affect Data from Multiple Mobile Devices” Ser. No. 14/144,413, filed Dec. 30, 2013 is also a continuation-in-part of U.S. patent application “Sporadic Collection of Mobile Affect Data” Ser. No. 14/064,136, filed Oct. 26, 2013, which claims the benefit of U.S. provisional patent applications “Sporadic Collection of Affect Data” Ser. No. 61/719,383, filed Oct. 27, 2012, “Optimizing Media Based on Mental State Analysis” Ser. No. 61/747,651, filed Dec. 31, 2012, “Collection of Affect Data from Multiple Mobile Devices” Ser. No. 61/747,810, filed Dec. 31, 2012, “Mental State Analysis Using Heart Rate Collection Based on Video Imagery” Ser. No. 61/793,761, filed Mar. 15, 2013, “Mental State Data Tagging for Data Collected from Multiple Sources” Ser. No. 61/790,461, filed Mar. 15, 2013, “Mental State Analysis Using Blink Rate” Ser. No. 61/789,038, filed Mar. 15, 2013, “Mental State Well Being Monitoring” Ser. No. 61/798,731, filed Mar. 15, 2013, and “Personal Emotional Profile Generation” Ser. No. 61/844,478, filed Jul. 10, 2013.

The patent application “Vehicular Cognitive Data Collection Using Multiple Devices” Ser. No. 15/886,275, filed Feb. 1, 2018 is also a continuation-in-part of U.S. patent application “Vehicle Manipulation using Occupant Image Analysis” Ser. No. 15/875,644, filed Jan. 19, 2018, which claims the benefit of U.S. provisional patent applications “Vehicle Manipulation using Occupant Image Analysis” Ser. No. 62/448,448, filed Jan. 20, 2017, “Image Analysis for Two-sided Data Hub” Ser. No. 62/469,591, filed Mar. 10, 2017, “Vehicle Artificial Intelligence Evaluation of Mental States” Ser. No. 62/503,485, filed May 9, 2017, “Image Analysis for Emotional Metric Generation” Ser. No. 62/524,606, filed Jun. 25, 2017, “Image Analysis and Representation for Emotional Metric Threshold Evaluation” Ser. No. 62/541,847, filed Aug. 7, 2017, “Multimodal Machine Learning for Emotion Metrics” Ser. No. 62/557,460, filed Sep. 12, 2017, “Speech Analysis for Cross-Language Mental State Identification” Ser. No. 62/593,449, filed Dec. 1, 2017, “Avatar Image Animation Using Translation Vectors” Ser. No. 62/593,440, filed Dec. 1, 2017, and “Directed Control Transfer for Autonomous Vehicles” Ser. No. 62/611,780, filed Dec. 29, 2017.

The patent application “Vehicle Manipulation Using Occupant Image Analysis” Ser. No. 15/875,644, filed Jan. 19, 2018 is also a continuation-in-part of U.S. patent application “Image Analysis in Support of Robotic Manipulation” Ser. No. 15/273,765, filed Sep. 23, 2016, which claims the benefit of U.S. provisional patent applications “Image Analysis In Support of Robotic Manipulation” Ser. No. 62/222,518, filed Sep. 23, 2015, “Analysis of Image Content with Associated Manipulation of Expression Presentation” Ser. No. 62/265,937, filed Dec. 10, 2015, “Image Analysis Using Sub-Sectional Component Evaluation To Augment Classifier Usage” Ser. No. 62/273,896, filed Dec. 31, 2015, “Analytics for Live Streaming Based on Image Analysis within a Shared Digital Environment” Ser. No. 62/301,558, filed Feb. 29, 2016, and “Deep Convolutional Neural Network Analysis of Images for Mental States” Ser. No. 62/370,421, filed Aug. 3, 2016.

The patent application “Image Analysis in Support of Robotic Manipulation” Ser. No. 15/273,765, filed Sep. 23, 2016 is a continuation-in-part of U.S. patent application “Mental State Event Definition Generation” Ser. No. 14/796,419, filed Jul. 10, 2015 which claims the benefit of U.S. provisional patent applications “Mental State Event Definition Generation” Ser. No. 62/023,800, filed Jul. 11, 2014, “Facial Tracking with Classifiers” Ser. No. 62/047,508, filed Sep. 8, 2014, “Semiconductor Based Mental State Analysis” Ser. No. 62/082,579, filed Nov. 20, 2014, and “Viewership Analysis Based On Facial Evaluation” Ser. No. 62/128,974, filed Mar. 5, 2015.

The patent application “Mental State Event Definition Generation” Ser. No. 14/796,419, filed Jul. 10, 2015 is also a continuation-in-part of U.S. patent application “Mental State Analysis Using Web Services” Ser. No. 13/153,745, filed Jun. 6, 2011, which claims the benefit of U.S. provisional patent applications “Mental State Analysis Through Web Based Indexing” Ser. No. 61/352,166, filed Jun. 7, 2010, “Measuring Affective Data for Web-Enabled Applications” Ser. No. 61/388,002, filed Sep. 30, 2010, “Sharing Affect Across a Social Network” Ser. No. 61/414,451, filed Nov. 17, 2010, “Using Affect Within a Gaming Context” Ser. No. 61/439,913, filed Feb. 6, 2011, “Recommendation and Visualization of Affect Responses to Videos” Ser. No. 61/447,089, filed Feb. 27, 2011, “Video Ranking Based on Affect” Ser. No. 61/447,464, filed Feb. 28, 2011, and “Baseline Face Analysis” Ser. No. 61/467,209, filed Mar. 24, 2011.

The patent application “Mental State Event Definition Generation” Ser. No. 14/796,419, filed Jul. 10, 2015 is also a continuation-in-part of U.S. patent application “Mental State Analysis Using an Application Programming Interface” Ser. No. 14/460,915, Aug. 15, 2014, which claims the benefit of U.S. provisional patent applications “Application Programming Interface for Mental State Analysis” Ser. No. 61/867,007, filed Aug. 16, 2013, “Mental State Analysis Using an Application Programming Interface” Ser. No. 61/924,252, filed Jan. 7, 2014, “Heart Rate Variability Evaluation for Mental State Analysis” Ser. No. 61/916,190, filed Dec. 14, 2013, “Mental State Analysis for Norm Generation” Ser. No. 61/927,481, filed Jan. 15, 2014, “Expression Analysis in Response to Mental State Express Request” Ser. No. 61/953,878, filed Mar. 16, 2014, “Background Analysis of Mental State Expressions” Ser. No. 61/972,314, filed Mar. 30, 2014, and “Mental State Event Definition Generation” Ser. No. 62/023,800, filed Jul. 11, 2014.

The patent application “Mental State Analysis Using an Application Programming Interface” Ser. No. 14/460,915, Aug. 15, 2014 is also a continuation-in-part of U.S. patent application “Mental State Analysis Using Web Services” Ser. No. 13/153,745, filed Jun. 6, 2011, which claims the benefit of U.S. provisional patent applications “Mental State Analysis Through Web Based Indexing” Ser. No. 61/352,166, filed Jun. 7, 2010, “Measuring Affective Data for Web-Enabled Applications” Ser. No. 61/388,002, filed Sep. 30, 2010, “Sharing Affect Across a Social Network” Ser. No. 61/414,451, filed Nov. 17, 2010, “Using Affect Within a Gaming Context” Ser. No. 61/439,913, filed Feb. 6, 2011, “Recommendation and Visualization of Affect Responses to Videos” Ser. No. 61/447,089, filed Feb. 27, 2011, “Video Ranking Based on Affect” Ser. No. 61/447,464, filed Feb. 28, 2011, and “Baseline Face Analysis” Ser. No. 61/467,209, filed Mar. 24, 2011.

Each of the foregoing applications is hereby incorporated by reference in its entirety.

This application relates generally to convolutional processing and more particularly to convolutional processing using a multilayered analysis engine.

Philosophers, physicians, and others have for centuries been fascinated by the human brain. The brain is used to process sensory information collected from one's surrounding environment in order to see whether an approaching human or animal is a friend or a foe, to smell smoke in the air, to hear an onrushing flood, to process pain, and to carry out other basic survival functions. The brain is further capable of understanding and producing speech, reading or writing text, solving complex problems, and creating art. For a variety of reasons, people have wondered how the brain works, how it can be used to learn and to store information, and how it makes an individual who she or he is. Some people have also wondered whether an artificial brain could be created, one that can learn, “think”, or act in human ways. For years, such musings were the stuff of science fiction or philosophical debate. In recent times, computer scientists, engineers, and mathematicians have designed and constructed systems that many consider loosely modeled on the human brain. These “neural networks” are constructed from networks of interconnected processing nodes called neurons. These neural networks execute a set of algorithms that mimic the processing of the human brain to solve a variety of problems.

Much as a child is taught to read, write, speak, or identify animals or objects, a neural network must be trained to solve a particular problem. The neural network can be trained by providing it with data related to the problem and the expected result. The neural network reads the data such as an image of a puppy, analyzes the data, and makes an inference about the data. If the neural network infers that the image contains a puppy, the neural network registers a success and moves on to the next data set or image. If the neural network draws the wrong inference, then it adjusts values, called weights, and tries again. The more training data that can be provided for training the neural network the better because the successful inference rate increases. The neural network is then given raw data and tries to make inferences about it based on the training. This seemingly simple approach to training is actually quite difficult. The training data is usually generated by humans, so that large training datasets are difficult to obtain or are simply unavailable. Further, an object or animal in an image can be difficult to identify. Dogs that look like cats, partially obscured objects such as keys which are somewhat concealed by other pocket contents, and other challenges, make the job of the neural network nearly impossible.

Disclosed neural networks represent processing techniques that are based on nodes that can process data, and arcs that can provide input data to nodes, transfer data between nodes, or provide resulting data from nodes to outputs. Neural networks differ from programs, codes, or applications in that while the applications remain static, the neural networks can be trained or can “learn”. Here, learning refers to the ability of the neural network, such as a deep learning neural network, to make adjustments to associated weights as more and more data is processed. The weights are adjusted such that the hit rate or success rate of a neural network to generate a correct inference regarding the data improves over time. Disclosed neural networks can be applied to many different types of tasks. The ability of the neural network to learn further enables the neural network to adapt to or track changes to the data. Thus, the disclosed neural network can provide an extremely powerful and beneficial way of processing data over traditional methods.

Disclosed embodiments provide for deep convolutional neural network computing. The convolutional computing is accomplished using a multilayered analysis engine. The multilayered analysis engine includes a deep learning network using a convolutional neural network (CNN). The multilayered analysis engine is used to analyze multiple images in a supervised or unsupervised learning process. Multiple images are provided to the multilayered engine, and the multilayered analysis engine is trained with those images. A subject image is then evaluated by the multilayered analysis engine by analyzing pixels within the subject image to identify a facial portion and to further identify a facial expression based on the facial portion. The results of the evaluation are output. The multilayered analysis engine is retrained using a second plurality of images.

A computer-implemented method for convolutional processing is disclosed comprising: initializing a computer for convolutional neural network processing; obtaining, using an imaging device, a plurality of images for processing on the computer for convolutional processing; training, on the computer, a multilayered analysis engine, using the plurality of images, wherein the multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer; evaluating a further image, using the multilayered analysis engine, wherein the evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion; and outputting a convolutional neural network analysis from the at least one output layer. Some embodiments include performing unsupervised learning as part of the training. In some embodiments, a layer from the multiple layers is fully connected. Some embodiments include performing supervised learning as part of the training by using a set of images, from the plurality of images, that have been labeled for cognitive states. Some embodiments include calculating a facial expression metric as part of the evaluating. Some embodiments include generating a cognitive state metric as part of the evaluating, based on the facial expression metric. And some embodiments include the cognitive state metric in the outputting. In some embodiments, a last layer within the multiple layers provides output indicative of cognitive state. And some embodiments include tuning the last layer within the multiple layers for a particular cognitive state.

Various features, aspects, and advantages of various embodiments will become more apparent from the following further description.

The following detailed description of certain embodiments may be understood by reference to the following figures wherein:

FIG. 1 is a flow diagram for convolutional computing using a multilayered analysis engine.

FIG. 2 is a flow diagram for providing cognitive state content.

FIG. 3 is an example illustrating facial data that can be used to generate a cognitive state metric.

FIG. 4 is a flow diagram representing deep convolutional processing image analysis.

FIG. 5 is a flow diagram representing training.

FIG. 6 is an example showing a pipeline for facial analysis layers.

FIG. 7 is an example illustrating a deep network for facial expression parsing.

FIG. 8 is an example illustrating a convolution neural network.

FIG. 9 is a system diagram for an interior of a vehicle.

FIG. 10 is a timeline with information tracks relating to cognitive states.

FIG. 11 shows cognitive state data with tags.

FIG. 12 shows example image and audio collection including multiple mobile devices.

FIG. 13 illustrates feature extraction for multiple faces.

FIG. 14 shows an example of livestreaming of social video and audio.

FIG. 15 is a flow diagram for detecting facial expressions.

FIG. 16 is a flow diagram for the large-scale clustering of facial events.

FIG. 17 shows unsupervised clustering of features and characterizations of cluster profiles.

FIG. 18A shows example tags embedded in a webpage.

FIG. 18B shows invoking tags to collect images.

FIG. 19 illustrates a bottleneck layer within a deep learning environment.

FIG. 20 shows data collection including multiple devices and locations.

FIG. 21 is a system diagram for convolutional computing using a multilayered analysis engine.

Convolutional neural network computing is difficult to train and configure for complex tasks. Cognitive analysis is a highly complex task. Understanding and evaluating cognitive states including moods, emotions, or mental states, require a nuanced evaluation of facial expressions or other cues that people generate. Techniques for a multilayered analysis engine used in convolutional computing are disclosed. The multilayered analysis can be used for cognitive analysis. Image processing is a critical element in cognitive state analysis. Cognitive state analysis is important in many areas. The understanding of cognitive states can be used in a variety of fields, such as improving marketing analysis, assessing the effectiveness of customer service and retail experiences, and evaluating the consumption of content such as movies and videos. Further, understanding of cognitive states can be applied to many areas of modern life, including, for example, assisting in the manipulation of autonomous or semi-autonomous vehicles. By understanding the one or more cognitive states being experienced by an occupant of a vehicle, a variety of vehicle manipulation steps can be taken.

For example, if a vehicle occupant is drowsy, inattentive, distracted, or impaired, control of an autonomous or semi-autonomous vehicle can be transferred from the occupant to the vehicle. A recommendation that the occupant take a break from operating a vehicle to rest, have a meal, and the like can be made. When a vehicle occupant is frustrated by traffic or construction delays, an alternative travel route can be recommended. Through evaluation of the various cognitive states, a cognitive state profile can be generated for the vehicle occupant. The cognitive state profile can be used to enable access by the person to a vehicle, to configure a vehicle to the person based on personal preferences, etc. Emotional state, mental state, cognitive state, and so on, are terms of art which may connote slight differences of emphasis, for example an emotional state of “happiness” vs. a cognitive state of “distractedness,” but at a high level, the terms can be used interchangeably. In fact, because the human mind of an individual is often difficult to understand, even for the individual, emotional, mental, and cognitive states may easily be overlapping and appropriately used in a general sense.

Discussed below and throughout, the analysis can include determining one or more metrics. Embodiments can include calculating a facial expression metric as part of the analyzing. The facial expression metric can include an intensity of a facial expression, a duration of a facial expression, an onset or decay of a facial expression, and so on. Embodiments include generating a cognitive state metric as part of the analyzing, based on the facial expression metric. Other metrics can be based on the audio data. The other metrics can be based on voice data within the audio data. In embodiments, the voice data enables audio vocalization metrics.

In embodiments, a multilayered analysis engine includes a computer initialized for convolutional processing of images of multiple faces. The result of the image processing can be output to a using application. The using application can include, for example, vehicle control. For example, an imaging device within a first vehicle is used to obtain a plurality of images of an occupant of the first vehicle. A multilayered analysis engine is trained on the computer initialized for convolutional processing using the plurality of images. The images can include images of a vehicle operator, one or more vehicle passengers, and so on. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers and one or more hidden layers. The multilayered analysis engine can include a pooling layer, a bottleneck layer, etc. A pooling layer can be an average pooling layer, a max pooling layer, or some other type of pooling layer. In embodiments, a convolution layer is used along with a dense layer and a max pooling layer within the multilayered analysis engine. In some embodiments, recurrent layers such as LSTM, GRU, or RNN layers are included in the multilayered analysis engine. The multilayered analysis engine is used for cognitive state analysis, where the cognitive state analysis can infer cognitive states based on the emotional content of the images. The cognitive states can include one or more of drowsiness, fatigue, distraction, impairment, fear, sadness, stress, happiness, anger, frustration, confusion, disappointment, hesitation, cognitive overload, focusing, engagement, attention, boredom, exploration, confidence, trust, delight, disgust, skepticism, doubt, satisfaction, excitement, laughter, calmness, curiosity, humor, depression, envy, sympathy, embarrassment, poignancy, or mirth. Further images are evaluated using the multilayered analysis engine. The further images include facial image data from one or more persons present in various settings.

FIG. 1 is a flow diagram for convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer.

The flow 100 includes initializing a computer 110 for convolutional processing. The computer can include one or more processors. The computer can be coupled to a handheld electronic device. The computer can include a server such as a local server, a blade server, a remote server, a cloud server, a mesh server, and the like. The flow 100 includes obtaining images 120, using an imaging device. The imaging device can include any of a variety of cameras, where the cameras can include a webcam, a video camera, a still camera, a thermal imager, a CCD device, a phone camera, a three-dimensional camera, a depth camera, a light field camera, multiple webcams used to show different views of a person, an infrared camera, or any other type of image capture apparatus that can allow captured data to be used in an electronic system. The images can include facial images of one or more people. The images can include different views of different faces comprising the facial image data.

The flow 100 includes training, on the computer initialized for convolutional processing, a multilayered analysis engine 130 using the plurality of images. The multilayered analysis engine includes multiple layers with one or more convolutional layers and one or more hidden layers, along with at least one output layer. The multilayered analysis engine is used for cognitive state analysis. The multilayered analysis engine can include a neural network (NN), a convolutional neural network (CNN), a deep neural network (DNN), and so on. A convolutional neural network, for example, can include an input layer, a convolutional layer, a pooling layer, a rectified linear units (ReLU) layer, a classification layer, an output layer, and so on. In embodiments, a layer from the multiple layers can be fully connected. In other embodiments, a pooling layer of the multilayered analysis engine can include the multilayered analysis engine and can further include a max pooling layer. The convolutional neural network can also include one or more of hidden layers or bottleneck layers. Cognitive states can be analyzed using the CNN, as discussed elsewhere. The training can encompass assigning weights 132 to inputs on one or more layers within the multilayered analysis engine. The weights can be assigned in order to enhance convergence, to improve accuracy, etc., of the convolutional processing. The assigning weights can be accomplished during a feed-forward pass through the multilayered analysis engine. In a feed-forward arrangement, information moves forward from the input nodes, through the hidden nodes, and on to the output nodes. The weights are updated during a backpropagation process through the multilayered analysis engine. When positive outcomes are encountered, the positive outcomes can be reinforced by changing weighting values within a model, algorithm, etc. Positive outcomes may result in increasing weighting values. Negative outcomes can also be considered, where weighting values may be reduced or otherwise adjusted.

The flow 100 includes performing supervised learning 134 as part of the training by using a set of images, from the plurality of images, that have been labeled for cognitive states. The supervised learning can include constructing a function that can map an input to an output based on training the function with known sample input-output pairs. That is, the function can be constructed to map a given input to a known output. The flow 100 includes performing unsupervised learning 136 as part of the training. The unsupervised learning can include inferring a function, where the function can be used to describe an unknown structure of the data. The dataset that can be used for the unstructured learning can be unlabeled. The flow 100 includes learning image descriptors 138, as part of the training, for emotional content. The image descriptors can be used to determine a connection among pixels in an image, such as an image that contains facial data. The descriptors can include color, texture, shape, motion, location, and so on. In embodiments, the image descriptors can be identified based on a temporal co-occurrence with an external stimulus. A temporal co-occurrence with an external stimulus can include a facial expression, cognitive state, emotional state, mental state, mood, and so on, that can occur at a time substantially similar to the time of the external stimulus. For example, an angry facial expression can occur when the occupant of the vehicle is cut off in traffic by a rude operator of another vehicle. The flow 100 further includes training an emotion classifier 140 for emotional content. An emotion classifier can be used to determine a given emotion based on collected and analyzed image data. The emotion classifier can be trained to identify emotional content. The emotion classifier can be used to infer a cognitive state based on emotional content within a face, where the cognitive state can include drowsiness, fatigue, distraction, impairment, and so on. In embodiments, the training of the multilayered analysis engine comprises deep learning. The deep learning can be used to identify cognitive states, mental states, emotional states, moods, and so on.

The flow 100 includes evaluating further images 150, using the multilayered analysis engine, where the further images include facial image data from one or more persons. The further images can be collected using a webcam, a video camera, and so on. The flow 100 includes outputting a convolutional neural network analysis 160 from at least one output layer of the neural network. The analysis can be used by another application. The using application can be running on the same or a separate computer that is configured for convolutional processing. The flow 100 includes inferring a cognitive state 170. A cognitive state can be based on emotional content, where the emotional content can be determined within a face that may be detected within the facial image data. The cognitive state data can be based on a wide range of emotional content. In embodiments, the cognitive state includes one or more of drowsiness, fatigue, distraction, impairment, sadness, stress, happiness, anger, frustration, confusion, disappointment, hesitation, cognitive overload, focusing, engagement, attention, boredom, exploration, confidence, trust, delight, disgust, skepticism, doubt, satisfaction, excitement, laughter, calmness, curiosity, humor, depression, envy, sympathy, embarrassment, poignancy, or mirth.

A last layer within the multiple layers of a multilayered analysis engine can provide an output indicative of cognitive state. The flow 100 further includes tuning the last layer within the multiple layers for a particular cognitive state 180. The turning the last layer can include adjusting weights for the last layer. The tuning the last layer can be used to discriminate among multiple, similar, cognitive states to better identify one or more cognitive states. The tuning the last layer can improve convergence of the multilayered analysis engine. The flow 100 includes retraining the multilayered analysis engine 190. The retraining can include adjusting weights of the various layers of the multilayered analysis engine. The retraining can update weights on a subset of layers within the multilayered analysis engine. In embodiments, the subset of layers includes a single layer within the multilayered analysis engine. The retraining can be accomplished using a feed-forward pass through the multilayered analysis engine, using backpropagation, and so on. The retraining can be based on using a second plurality of images 192. The second plurality of images can include images of the same or different individuals. Various steps in the flow 100 may be changed in order, repeated, omitted, or the like without departing from the disclosed concepts. Various embodiments of the flow 100 can be included in a computer program product embodied in a non-transitory computer readable medium that includes code executable by one or more processors. Various embodiments of the flow 100, or portions thereof, can be included on a semiconductor chip and implemented in special purpose logic, programmable logic, and so on.

FIG. 2 is a flow diagram for providing cognitive state content. Discussed throughout, a convolutional processor can be used to analyze image data, including facial image data, to determine the presence or absence of a cognitive state, an intensity or duration of the cognitive state, a cognitive state exceeding a threshold, and so on. For example, images can be obtained from a first set of participants within a plurality of participants involved in an interactive digital environment, and cognitive state content can be analyzed within the images. The flow 200 includes calculating a facial expression metric 210 as part of the analyzing. Recall that the analyzing includes analyzing cognitive state content (discussed below) within the plurality of images for the first set of participants within the plurality of participants. The facial expression can include a smile, a frown, a smirk, a neutral expression, and so on. The facial expression can be determined using a system such as the facial action coding system (FACS) described below. A facial expression such as a smile can be coded based on an intensity, a speed of onset or decay, a duration, etc. A metric associated with the facial expression can be calculated, where the metric can include a value, a percentage, and the like. The metric can be calculated based on exceeding a threshold. Other metrics, such as audio vocalization metrics can also be calculated. The vocalization metric can include a laughter metric, a groan metric, a sigh metric, a yawn metric, and so on.

The flow 200 includes calculating a cognitive state metric 220 as part of the analyzing, based on the facial expression metric. Discussed throughout, a cognitive state metric, like a facial expression metric or vocalization metric, can be based on an intensity, a presence or absence, a duration, onset, decay, and so on. The cognitive state metric can be based on a variety of cognitive states such as engagement, distractedness, focus, concentration, ennui or boredom, drowsiness, and the like. The flow 200 includes providing results of analyzing cognitive state content of images to participants 230 within a shared digital environment. The participants can include a first set of participants within the plurality of participants, a second set of participants within the plurality of participants, and so on. The participants can include presenters, viewers, gamers, and further participants assuming roles within the shared digital environment. The providing can include providing an emoji, an animation such as an animated GIF, a laugh track, etc. The flow 200 includes including the cognitive state metric 232 in the providing. The cognitive state metric can be provided to a number of participants, to presenters only, to viewers only, and so on. The cognitive state metric can include an ideogram or a pictogram (e.g. emoji), a value, an animation, a graph of the cognitive state metric plotted versus time, etc. Various embodiments of the flow 200 can be included in a computer program product embodied in a non-transitory computer readable medium that includes code executable by one or more processors. Various embodiments of the flow 200, or portions thereof, can be included on a semiconductor chip and implemented in special purpose logic, programmable logic, and so on.

FIG. 3 is an example illustrating facial data that can be used to generate a cognitive state metric. A cognitive state metric can be based on a cognitive state such as engagement, distractedness, ennui, drowsiness, and so on. One or more cognitive state metrics can be determined by analyzing cognitive state content within image data, audio data, and so on, obtained from images of participants of a shared digital environment. The metrics can be based on livestreaming analytics within a shared digital environment. FIG. 3 includes three charts, charts 310, 312, and 314. Each chart has a horizontal axis of time, and a vertical axis of an engagement level, which may be derived from cognitive state data. In other embodiments, cognitive state data or other data derived from cognitive state data may be used to generate cognitive state metrics, such as measures of happiness, inattentiveness, concentration, and so on. Each bar on the chart may represent a time window comprising a fixed unit of time, such as one minute. In chart 310, until time t1, the engagement level is at 92%, indicating that the user is mostly focused on the displayed content. After time t1, the next bar indicates a very low engagement level because at some point during that time window, the user left the area. In the subsequent time windows, the engagement level is zero, as the individual is no longer present.

In chart 312, the individual remains present in front of the rendered content, but for a portion of the video, he frequently looks away. As can be seen in the chart 312, up until time t2, the engagement level is sporadic, fluctuating between low and midrange levels. After time t2, the engagement level increases. In such an embodiment where digital media content is modified based on viewership, a chart such as 312 indicates that the ending of the video is engaging to the individual, while earlier in the video, before time t2, the video was not as engaging. Thus, in embodiments, the modification includes shortening the video by deleting and/or shortening scenes of the video prior to time t2, in order to better hold the individual's attention and interest.

In chart 314, the individual remains present in front of the rendered content, but for a portion of the video, he is frequently looking away by averting his gaze away from the screen that is presenting the media content. As can be seen in chart 314, up until time t3, the engagement level is relatively high, indicating a high level of focus by the individual on the media content. After time t3, the engagement level significantly decreases. Each detected engagement level may be considered cognitive state data. In order to generate a cognitive state metric based on a chart such as 314, the cognitive state data may be processed in any appropriate and desired fashion.

For example, groups of three sequential engagement levels may be averaged to produce cognitive state metrics for a plurality of time periods. As another example, all of the engagement levels for a given time period may be summed and divided by the number of engagement levels that are below 50% in order to determine a cumulative cognitive state metric. For example, in chart 310, a cumulative cognitive state metric may be determined by summing all of the engagement levels (560) and dividing by the number of engagement levels below 50% (ten), resulting in a cumulative cognitive state metric of 560/10 or 56. For chart 310, a cumulative cognitive state metric may be determined by summing all of the engagement levels (543.1) and dividing by the number of engagement levels below 50% (ten), resulting in a cumulative cognitive state metric of 543.1/10 or 54.31. For chart 314, a cumulative cognitive state metric may be determined by summing all of the engagement levels (560) and dividing by the number of engagement levels below 50% (ten in chart 314), resulting in a cumulative cognitive state metric of 56. Thus, if chart 310 has a cumulative cognitive state metric of 56, chart 312 has a metric of 54.31, and chart 314 has a metric of 56, it may be determined that charts 310 and 314 indicate roughly equal levels of engagement while chart 312 indicates slightly lower engagement than that shown by charts 310 and 314. As further examples, if a user is 100% engaged for 8 of 16 sample periods and 49% engaged for the remaining eight sample periods, the cumulative cognitive state metric may be calculated as 100, indicating more engagement than is shown in charts 310, 312, and 314. However, if a user is only 80% engaged for 4 of 16 sample periods and 0% engaged for the remaining 12 sample periods, the cumulative cognitive state metric may be calculated as 26.67, indicating less engagement than is shown in charts 310, 312, and 314. Although only a selection of cognitive state metrics is explicitly discussed herein, it will be understood after reviewing this application in its entirety that any number of different cognitive state metrics may be used.

FIG. 4 is a flow diagram representing deep convolutional processing image analysis. Convolutional computing can use a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The flow 400 shows a computer-implemented method for image analysis. The flow 400 includes initializing a convolutional computer 402. A convolutional computer can either be specialized hardware designed specifically for neural network convolution, or it can comprise unique software that enables a generic computer to operate as a specialized convolutional machine. The convolutional computer may exist as dedicated hardware, or it may exist as part of a networked structure, such as a supercomputer, a supercomputer cluster, a cloud-based system, a server-based system, a distributed computer network, and the like. The flow 400 includes obtaining a plurality of images 410. Each of the plurality of images can include at least one human face. In embodiments, each image includes metadata. The metadata can include information about each face that is entered by human coders. The metadata can include a perceived facial expression and/or cognitive state. The metadata can further include demographic information such as an age range, gender, and/or ethnicity. Since different demographic groups might register emotions in different ways, the demographic information can be used to further enhance the output results of the multilayered analysis engine. For example, some demographic groups might not smile as frequently or as intensely as others. Thus, a compensation can be applied in these circumstances when analyzing smiles to effectively normalize the results. A similar approach can be applied to other facial expressions and/or cognitive states.

The flow 400 includes training a multilayered analysis engine 420 using the plurality of images, wherein the multilayered analysis engine comprises multiple layers that include one or more convolutional layers and one or more hidden layers, and wherein the multilayered analysis engine is used for emotional analysis. Hidden layers are layers within the multilayered analysis engine with outputs that are not externally exposed. The output of hidden layers feeds another layer, but the output of the hidden layer is not directly observable. The training can include submitting multiple images to the multilayered analysis engine. In a supervised learning scenario, the multilayered analysis engine makes an assessment of the facial expression and/or cognitive state and compares its output to the human-coded assessment in the metadata. Various parameters such as weights between the layers can be adjusted until a majority of the input images are correctly classified by the multilayered analysis engine. Alternate embodiments can be implemented wherein stopping criteria is used. A desired target or accuracy is selected and images that have been labeled are analyzed. Once the target cognitive states or facial expressions identified by the neural network match those which have been labeled, then the learning can be stopped. In some cases, stopping criteria can be based on the number of images that are matched to the number of re-learning steps used, the decrease in error rates, the reduction of error rates to a predefined limit, the number of back propagation operations performed, and so on.

Thus, the flow 400 can further include assigning weights 424. The assignment of weights can be influenced by updating the weights during backpropagation 426. Back propagation can include calculation of a loss function gradient which is used to update the values of the weights as part of a supervised learning process. The assignment of weights can be selected to emphasize facial features 428, such as eyes, mouth, nose, eyelids, eyebrows, and/or chin. In other embodiments, the input images are not associated with metadata pertaining to facial expressions and/or cognitive states. In such cases, the multilayered analysis engine is trained using an unsupervised learning process.

The flow 400 includes evaluating a further image 440 using the multilayered analysis engine, wherein the evaluating includes analyzing pixels within the further image to identify a facial portion 422 and identifying a facial expression 446 based on the facial portion. The facial expression can include a smile, frown, laugh, or an expression of surprise, concern, confusion, and/or anger, among others. The analyzing of pixels for identifying a facial portion 422 can include identifying a face contour, as well as locating facial features such as eyes, nose, mouth, chin, and cheekbones. The further image 440 is a subject image that is to be analyzed by the multilayered analysis engine. Thus, once the multilayered analysis engine is trained, a subject image can be input to the multilayered analysis engine, and the multilayered analysis engine can analyze the subject image to determine a facial expression 446 and/or an output indicative of a cognitive state 450. A facial expression can correlate to more than one cognitive state, depending on the circumstances. For example, a smile can indicate happiness in many situations. However, in some cases, a person might smile while experiencing another cognitive state, like embarrassment. The “happy” smile might have slightly different attributes than the “embarrassed” smile. For example, the lip corners can be pulled higher in a “happy” smile than in an “embarrassed” smile. Through the training of the multilayered analysis engine 420, the multilayered analysis engine can learn the difference between the variants of a facial expression (e.g. a smile) to provide an output indicative of cognitive state 450.

The flow 400 includes analyzing an emotion 442. The emotion can be a representation of how the subject person is feeling at the time of image acquisition. The emotion analysis can be based on facial features and can include the use of action units (AUs). Such AUs can include, but are not limited to, brow lowerer, nose wrinkler, and mouth stretcher, just to name a few. In practice, many more AUs can be examined during emotion analysis 442. The flow 400 includes inferring a cognitive state based on emotional content within a face associated with the facial portion 444. The emotional content can include, but is not limited to, facial expressions such as smiles, smirks, and frowns. Emotional content can also include actions such as lip biting, eye shifting, and head tilting. Furthermore, external features such as tears on a face can be part of the emotional content. For example, detecting the presence of tears can be used in determining an expression/cognitive state of sorrow. However, in some instances, tears can also signify a cognitive state of extreme joy. The multilayered analysis engine can examine other factors in conjunction with the presence of tears to distinguish between the expressions of sorrow and joy.

The flow 400 includes tuning one or more layers 438 within the multiple layers for a particular cognitive state. In some embodiments, the tuning is for the last layer within the multiple layers where that last layer is tuned for identifying a particular cognitive state. In other embodiments, multiple layers are tuned. The multilayered analysis engine can include many layers. In embodiments, tuning the last layer is used to adjust the output so that the cognitive state and/or facial expressions provided by the multilayered analysis engine agree with the images used to train the multilayered analysis engine. For example, if images used for training contain facial expressions indicative of joy, but the output provided by the multilayered analysis engine does not indicate joy in a majority of the cases, then the last layer can be tuned so that the output provided by the multilayered analysis engine will indicate joy in a majority of the cases. The tuning can include adjusting weights, constants, or functions within and/or input to the last layer. Furthermore, other tuning techniques can be employed, including learning from previous layers. In addition, later layers can be tuned to learn different expressions or further expressions so that other cognitive states or facial expressions are identified by the neural network.

The flow 400 includes identifying boundaries of the face 430. Identifying the existence of a face within an image can be accomplished in a variety of ways, including, but not limited to, utilizing a histogram-of-oriented-gradient (HoG)-based object detectors. The flow 400 includes identifying landmarks of the face 432. The landmarks are points of interest within a face. These can include, but are not limited to, the right eye, left eye, nose base, and lip corners. The flow 400 includes extracting features of the face 434. The features can include, but are not limited to, eyes, eyebrows, eyelids, lips, lip corners, chin, cheeks, teeth, and dimples.

The flow 400 can include various types of face normalization 436 such as rotating, resizing, contrast adjustment, brightness adjustment, cropping, and so on. One or more of these normalization processes can be executed on faces within the plurality of images. The normalization steps can be performed on images, videos, or frames within a video. In embodiments, the image is rotated to a fixed orientation by an input layer of the multilayered analysis engine. For example, a face that is tilted at a 30-degree angle can be rotated such that it is oriented vertically, so that the mouth is directly below the nose of the face. In this way, the subsequent layers of the multilayered analysis engine work with a consistent image orientation.

As part of the training, the flow 400 includes training an emotion classifier for emotional content. The emotion classifier can include one or more of sadness, stress, happiness, anger, frustration, confusion, disappointment, hesitation, cognitive overload, focusing, engagement, attention, boredom, exploration, confidence, trust, delight, disgust, skepticism, doubt, satisfaction, excitement, laughter, calmness, shock, surprise, fear, curiosity, humor, sadness, poignancy, or mirth. In embodiments, the multilayered analysis engine is trained for a specific emotion such as shock. For example, in an application for determining the effectiveness of scary scenes in a horror movie, the multilayered analysis engine can be specifically trained to identify a facial expression of shock, corresponding to a cognitive state of surprise combined with fear. The horror movie is then shown to a test audience, where one or more cameras obtain images of the audience as the movie is being viewed. Facial images are acquired at a predetermined time after the presentation of a scary scene. The images can be acquired at a time ranging from about 300 milliseconds to about 700 milliseconds after presentation of the scary scene. This allows a viewer sufficient processing time to react to the scene, but is not so long that the viewer is no longer expressing their initial reaction.

In the flow 400, the training of the multilayered analysis engine comprises deep learning. Deep learning is a type of machine learning utilizing neural networks. In general, it is difficult for a computer to interpret the meaning of raw sensory input data, such as digital images that are represented as an array of pixels. Converting from an array or subset of pixels to identification of an object within the image, such as a human face, is very complicated. Direct evaluation of this mapping is computationally impractical to solve directly. However, embodiments disclosed herein comprise a multilayered analysis engine that utilizes deep learning. The multilayered analysis engine can determine features within an image by dividing the highly complex mapping into a series of more simple mappings, each processed by a different layer of the multilayered analysis engine. The input image is presented to an input layer, which performs initial processing on the image. Then one or more hidden layers extract features from the image. In embodiments, the outputs of the hidden layers are not directly observable. The hidden layers can provide evaluation of cognitive states or facial expressions without specific interpretation or labels being provided. The outputs of the hidden layers can, however, be used by further layers within the convolutional neural network to perform the cognitive state or facial expression analysis.

When an image is input to the multilayered analysis engine, the input layer can be used to identify edges by comparing the brightness of neighboring pixels or by using other edge detection process. The edges can then be input to a subsequent hidden layer, which can then extract features such as corners. The process continues with additional hidden layers, each additional layer performing additional operations, and culminating with an output layer that produces a result which includes a facial expression and/or cognitive state. Thus, the deep learning network provides an improvement in automated detection of facial expressions and/or cognitive states, enabling new and exciting applications such as large-scale evaluation of emotional response.

In the flow 400, the multilayered analysis engine comprises a convolutional neural network. Convolutional neural networks (CNNs) share many properties with ordinary neural networks. For example, they both include neurons that have learnable weights and biases. Each node/neuron receives some inputs and performs a function that determines if the node/neuron “fires” and generates an output. However, CNNs are well suited for inputs that are images, allowing for certain optimizations to be incorporated into the architecture of the CNN. CNNs make the forward function more efficient to implement and they improve the performance of image analysis. The flow 400 further includes inferring a cognitive state based on emotional content within a face detected within the facial image data, wherein the cognitive state includes of one or more of drowsiness, fatigue, distraction, impairment, sadness, stress, happiness, anger, frustration, confusion, disappointment, hesitation, cognitive overload, focusing, engagement, attention, boredom, exploration, confidence, trust, delight, disgust, skepticism, doubt, satisfaction, excitement, laughter, calmness, curiosity, humor, depression, envy, sympathy, embarrassment, poignancy, or mirth. Various steps in the flow 400 may be changed in order, repeated, omitted, or the like without departing from the disclosed concepts. Various embodiments of the flow 400 can be included in a computer program product embodied in a non-transitory computer readable medium that includes code executable by one or more processors. Various embodiments of the flow 400, or portions thereof, can be included on a semiconductor chip and implemented in special purpose logic, programmable logic, and so on.

FIG. 5 is a flow diagram representing training. A multilayered analysis engine can be trained for convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The flow 500 includes training a multilayered analysis engine using a first plurality of images 510. The flow 500 includes assigning weights to inputs 520. The weights can be applied to inputs to the layers that comprise the multilayered analysis engine. In some embodiments, the weights are assigned an initial value that update during the training of the multilayered analysis engine based on processes such as backpropagation. In embodiments, the flow 500 includes performing supervised learning 530 as part of the training by using a set of images from the plurality of images that have been labeled for cognitive states. In other embodiments, the flow 500 includes performing unsupervised learning 540 as part of the training.

As part of the training, the flow 500 includes learning image descriptors 550 for emotional content. The image descriptors can include features within an image such as those represented by action units (AUs). The descriptors can include, but are not limited to, features such as a raised eyebrow, a wink of one eye, or a smirk. In the flow 500, the image descriptors are identified based on a temporal co-occurrence with an external stimulus. The external stimulus can include media content such as an advertisement, a scene from a movie, or an audio clip. Additionally, the external stimulus can include a live event such as a siren, a thunder clap, or a flashing light that occurs in the room where the subject is located. The flow 500 includes training emotional classifiers 560. By analyzing multiple training images, the multilayered analysis engine can learn that lip corners pulled down in conjunction with lowered brows may be indicative of a cognitive state of disappointment. As more and more images are reviewed, the multilayered analysis engine generally becomes better equipped to analyze cognitive state and/or facial expressions.

The flow 500 includes retraining the multilayered analysis engine using a second plurality of images 570. In some embodiments, once an initial training session has been completed, the retraining occurs using images of a specific subset of emotions. For example, the second plurality of images can focus exclusively on fear, shock, and surprise. The second plurality of images can be tailored to the emotions of interest for the users of the multilayered analysis engine. In the flow 500, the retraining updates weights which are applied to a subset of layers within the multilayered analysis engine. In embodiments, the subset of layers is a single layer within the multilayered analysis engine. Additionally, the flow 500 can include the use of deep learning 512 to accomplish the training. Various steps in the flow 500 may be changed in order, repeated, omitted, or the like without departing from the disclosed concepts. Various embodiments of the flow 500 can be included in a computer program product embodied in a non-transitory computer readable medium that includes code executable by one or more processors. Various embodiments of the flow 500, or portions thereof, can be included on a semiconductor chip and implemented in special purpose logic, programmable logic, and so on.

FIG. 6 is an example showing a pipeline for facial analysis layers. A pipeline of facial analysis layers can be applied to convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The example 600 includes an input layer 610. The input layer 610 receives image data. The image data can be input in a variety of formats, such as JPEG, TIFF, BMP, and GIF. Compressed image formats can be decompressed into arrays of pixels, wherein each pixel can include an RGB tuple. The input layer 610 can then perform processing such as identifying boundaries of the face, identifying landmarks of the face, extracting features of the face, and/or rotating a face within the plurality of images. The output of the input layer can then be input to a convolutional layer 620. The convolutional layer 620 can represent a convolutional neural network and can contain a plurality of hidden layers. A layer from the multiple layers can be fully connected. The convolutional layer 620 can reduce the amount of data feeding into a fully connected layer 630. The fully connected layer processes each pixel/data point from the convolutional layer 620. A last layer within the multiple layers can provide output indicative of a certain cognitive state. The last layer is the final classification layer 640. The output of the final classification layer 640 can be indicative of the cognitive states of faces within the images that are provided to input layer 610.

FIG. 7 is an example 700 illustrating a deep network for facial expression parsing. Facial expression parsing can be applied to convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The evaluating provides a cognitive state analysis. A first layer 710 of the deep network is comprised of a plurality of nodes 712. Each of nodes 712 serves as a neuron within a neural network. The first layer can receive data from an input layer. The output of the first layer 710 feeds to the next layer 720. The layer 720 further comprises a plurality of nodes 722. A weight 714 adjusts the output of the first layer 710 which is being input to the layer 720. In embodiments, the layer 720 is a hidden layer. The output of the layer 720 feeds to a subsequent layer 730. That layer 730 further comprises a plurality of nodes 732. A weight 724 adjusts the output of the second layer 720 which is being input to the third layer 730. In embodiments, the third layer 730 is also a hidden layer. The output of the third layer 730 feeds to a fourth layer 740 which further comprises a plurality of nodes 742. A weight 734 adjusts the output of the third layer 730 which is being input to the fourth layer 740. The fourth layer 740 can be a final layer, providing a facial expression and/or cognitive state as its output. The facial expression can be identified using a hidden layer from the one or more hidden layers. The weights can be provided on inputs to the multiple layers to emphasize certain facial features within the face. The training can comprise assigning weights to inputs on one or more layers within the multilayered analysis engine. In embodiments, one or more of the weights (714, 724, and/or 734) can be adjusted or updated during training. The assigning weights can be accomplished during a feed-forward pass through the multilayered analysis engine. In a feed-forward arrangement, the information moves forward from the input nodes through the hidden nodes and on to the output nodes. Additionally, the weights can be updated during a backpropagation process through the multilayered analysis engine.

FIG. 8 is an example illustrating a convolution neural network (CNN). A convolutional neural network such as the network 800 can be used for deep learning, where the deep learning can be applied to convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The convolutional neural network can be applied to such tasks as cognitive state analysis, mental state analysis, mood analysis, emotional state analysis, and so on. Cognitive state data can include mental processes, where the mental processes can include attention, creativity, memory, perception, problem solving, thinking, use of language, and the like.

Cognitive analysis is a very complex task. Understanding and evaluating moods, emotions, mental states, or cognitive states, requires a nuanced evaluation of facial expressions or other verbal and nonverbal cues that people generate. Cognitive state analysis is important in many areas such as research, psychology, business, intelligence, law enforcement, and so on. The understanding of cognitive states can be useful for a variety of business purposes such as improving marketing analysis, assessing the effectiveness of customer service interactions and retail experiences, and evaluating the consumption of content such as movies and videos. Identifying points of frustration in a customer transaction can allow a company to take action to address the causes of the frustration. By streamlining processes, key performance areas such as customer satisfaction and customer transaction throughput can be improved, resulting in increased sales and revenues. In a content scenario, producing compelling content that achieves the desired effect (e.g. fear, shock, laughter, etc.) can boost ticket sales and/or advertising revenue. If a movie studio is producing a horror movie, it is desirable to know if the scary scenes in the movie are achieving the desired effect. By conducting tests in sample audiences, and analyzing faces in the audience, a computer-implemented method and system can process thousands of faces to assess the cognitive state at the time of the scary scenes. In many ways, such an analysis can be more effective than surveys that ask audience members questions, since audience members may consciously or subconsciously change answers based on peer pressure or other factors. However, spontaneous facial expressions can be more difficult to conceal. Thus, by analyzing facial expressions en masse in real time, important information regarding the general cognitive state of the audience can be obtained.

Analysis of facial expressions is also a complex task. Image data, where the image data can include facial data, can be analyzed to identify a range of facial expressions. The facial expressions can include a smile, frown, smirk, and so on. The image data and facial data can be processed to identify the facial expressions. The processing can include analysis of expression data, action units, gestures, mental states, cognitive states, physiological data, and so on. Facial data as contained in the raw video data can include information on one or more of action units, head gestures, smiles, brow furrows, squints, lowered eyebrows, raised eyebrows, attention, and the like. The action units can be used to identify smiles, frowns, and other facial indicators of expressions. Gestures can also be identified, and can include a head tilt to the side, a forward lean, a smile, a frown, as well as many other gestures. Other types of data including physiological data can be collected, where the physiological data can be obtained using a camera or other image capture device, without contacting the person or persons. Respiration, heart rate, heart rate variability, perspiration, temperature, and other physiological indicators of cognitive state can be determined by analyzing the images and video data.

Deep learning is a branch of machine learning which seeks to imitate in software the activity which takes place in layers of neurons in the neocortex of the human brain. This imitative activity can enable software to “learn” to recognize and identify patterns in data, where the data can include digital forms of images, sounds, and so on. The deep learning software is used to simulate the large array of neurons of the neocortex. This simulated neocortex, or artificial neural network, can be implemented using mathematical formulas that are evaluated on processors. With the proliferating capabilities of the processors, increasing numbers of layers of the artificial neural network can be processed.

Deep learning applications include processing of image data, audio data, and so on. Image data applications include image recognition, facial recognition, etc. Image data applications can include differentiating dogs from cats, identifying different human faces, and the like. The image data applications can include identifying cognitive states, moods, mental states, emotional states, and so on, from the facial expressions of the faces that are identified. Audio data applications can include analyzing audio such as ambient room sounds, physiological sounds such as breathing or coughing, noises made by an individual such as tapping and drumming, voices, and so on. The voice data applications can include analyzing a voice for timbre, prosody, vocal register, vocal resonance, pitch, volume, speech rate, or language content. The voice data analysis can be used to determine one or more cognitive states, moods, mental states, emotional states, etc.

The artificial neural network, such as a convolutional neural network which forms the basis for deep learning, is based on layers. The layers can include an input layer, a convolutional layer, a fully connected layer, a classification layer, and so on. The input layer can receive input data such as image data, where the image data can include a variety of formats including pixel formats. The input layer can then perform processing tasks such as identifying boundaries of the face, identifying landmarks of the face, extracting features of the face, and/or rotating a face within the plurality of images. The convolutional layer can represent an artificial neural network such as a convolutional neural network. A convolutional neural network can contain a plurality of hidden layers. A convolutional layer can reduce the amount of data feeding into a fully connected layer. The fully connected layer processes each pixel/data point from the convolutional layer. A last layer within the multiple layers can provide output indicative of cognitive state. The last layer of the convolutional neural network can be the final classification layer. The output of the final classification layer can be indicative of the cognitive states of faces within the images that are provided to the input layer.

Deep networks including deep convolutional neural networks can be used for facial expression parsing. A first layer of the deep network includes multiple nodes, where each node represents a neuron within a neural network. The first layer can receive data from an input layer. The output of the first layer can feed to a second layer, where the latter layer also includes multiple nodes. A weight can be used to adjust the output of the first layer which is being input to the second layer. Some layers in the convolutional neural network can be hidden layers. The output of the second layer can feed to a third layer. The third layer can also include multiple nodes. A weight can adjust the output of the second layer which is being input to the third layer. The third layer may be a hidden layer. Outputs of a given layer can be fed to the next layer. Weights adjust the output of one layer as it is fed to the next layer. When the final layer is reached, the output of the final layer can be a facial expression, a cognitive state, a mental state, a characteristic of a voice, and so on. The facial expression can be identified using a hidden layer from the one or more hidden layers. The weights can be provided on inputs to the multiple layers to emphasize certain facial features within the face. The convolutional neural network can be trained to identify facial expressions, voice characteristics, etc. The training can include assigning weights to inputs on one or more layers within the multilayered analysis engine. One or more of the weights can be adjusted or updated during training. The assigning weights can be accomplished during a feed-forward pass through the multilayered neural network. In a feed-forward arrangement, the information moves forward from the input nodes, through the hidden nodes, and on to the output nodes. Additionally, the weights can be updated during a backpropagation process through the multilayered analysis engine.

Returning to the figure, FIG. 8 is an example showing a convolutional neural network 800. The convolutional neural network can be used for deep learning, where the deep learning can be applied to avatar image animation using translation vectors. The deep learning system can be accomplished using a convolutional neural network or other techniques. The deep learning can perform facial recognition and analysis tasks. The network includes an input layer 810. The input layer 810 receives image data. The image data can be input in a variety of formats, such as JPEG, TIFF, BMP, and GIF. Compressed image formats can be decompressed into arrays of pixels, wherein each pixel can include an RGB tuple. The input layer 810 can then perform processing such as identifying boundaries of the face, identifying landmarks of the face, extracting features of the face, and/or rotating a face within the plurality of images.

The network includes a collection of intermediate layers 820. The multilayered analysis engine can include a convolutional neural network. Thus, the intermediate layers can include a convolutional layer 822. The convolutional layer 822 can include multiple sublayers, including hidden layers within it. The output of the convolutional layer 822 feeds into a pooling layer 824. The pooling layer 824 performs a data reduction, which makes the overall computation more efficient. Thus, the pooling layer reduces the spatial size of the image representation to reduce the number of parameters and computations in the network. In some embodiments, the pooling layer is implemented using filters of size 2×2, applied with a stride of two samples for every depth slice along both width and height, resulting in a reduction of 75-percent of the downstream node activations. The multilayered analysis engine can further include a max pooling layer as part of pooling layer 824. Thus, in embodiments, the pooling layer is a max pooling layer, in which the output of the filters is based on a maximum of the inputs. For example, with a 2×2 filter, the output is based on a maximum value from the four input values. In other embodiments, the pooling layer is an average pooling layer or L2-norm pooling layer. Various other pooling schemes are possible.

The intermediate layers can include a Rectified Linear Units (RELU) layer 826. The output of the pooling layer 824 can be input to the RELU layer 826. In embodiments, the RELU layer implements an activation function such as ƒ(x)−max(0,x), thus providing an activation with a threshold at zero. In some embodiments, the RELU layer 826 is a leaky RELU layer. In this case, instead of the activation function providing zero when x<0, a small negative slope is used, resulting in an activation function such as ƒ(x)=1(x<0)(αx)+1(x>=0)(x). This can reduce the risk of “dying RELU” syndrome, where portions of the network can be “dead” with nodes/neurons that do not activate across the training dataset. The image analysis can comprise training a multilayered analysis engine using the plurality of images, wherein the multilayered analysis engine can comprise multiple layers that include one or more convolutional layers 822 and one or more hidden layers, and wherein the multilayered analysis engine can be used for emotional analysis.

The example 800 includes a fully connected layer 830. The fully connected layer 830 processes each pixel/data point from the output of the collection of intermediate layers 820. The fully connected layer 830 takes all neurons in the previous layer and connects them to every single neuron it has. The output of the fully connected layer 830 provides input to a classification layer 840. The output of the classification layer 840 provides a facial expression and/or cognitive state. Thus, a multilayered analysis engine such as the one depicted in FIG. 8 processes image data using weights, models the way the human visual cortex performs object recognition and learning, and effectively analyzes image data to infer facial expressions and cognitive states.

Machine learning for generating parameters, analyzing data such as facial data and audio data, and so on, can be based on a variety of computational techniques. Generally, machine learning can be used for constructing algorithms and models. The constructed algorithms, when executed, can be used to make a range of predictions relating to data. The predictions can include whether an object in an image is a face, a box, or a puppy, whether a voice is female, male, or robotic, whether a message is legitimate email or a “spam” message, and so on. The data can include unstructured data and can be of large quantity. The algorithms that can be generated by machine learning techniques are particularly useful to data analysis because the instructions that comprise the data analysis technique do not need to be static. Instead, the machine learning algorithm or model, generated by the machine learning technique, can adapt. Adaptation of the learning algorithm can be based on a range of criteria such as success rate, failure rate, and so on. A successful algorithm is one that can adapt—or learn—as more data is presented to the algorithm. Initially, an algorithm can be “trained” by presenting it with a set of known data (supervised learning). Another approach, called unsupervised learning, can be used to identify trends and patterns within data. Unsupervised learning is not trained using known data prior to data analysis.

Reinforced learning is an approach to machine learning that is inspired by behaviorist psychology. The underlying premise of reinforced learning (also called reinforcement learning) is that software agents can take actions in an environment. The actions taken by the agents should maximize a goal such as a “cumulative reward”. A software agent is a computer program that acts on behalf of a user or other program. The software agent is implied to have the authority to act on behalf of the user or program. The actions taken are decided by action selection to determine what to do next. In machine learning, the environment in which the agents act can be formulated as a Markov decision process (MDP). The MDPs provide a mathematical framework for modeling of decision making in environments where the outcomes can be partly random (stochastic) and partly under the control of the decision maker. Dynamic programming techniques can be used for reinforced learning algorithms. Reinforced learning is different from supervised learning in that correct input/output pairs are not presented, and suboptimal actions are not explicitly corrected. Rather, on-line or computational performance is the focus. On-line performance includes finding a balance between exploration of new (uncharted) territory or spaces and exploitation of current knowledge. That is, there is a tradeoff between exploration and exploitation.

Machine learning based on reinforced learning adjusts or learns based on learning an action, a combination of actions, and so on. An outcome results from taking an action. Thus, the learning model, algorithm, etc., learns from the outcomes that result from taking the action or combination of actions. The reinforced learning can include identifying positive outcomes, where the positive outcomes are used to adjust the learning models, algorithms, and so on. A positive outcome can be dependent on a context. When the outcome is based on a mood, emotional state, mental state, cognitive state, etc., of an individual, then a positive mood, emotion, mental state, or cognitive state can be used to adjust the model and algorithm. Positive outcomes can include the person being more engaged, where engagement is based on affect, the person spending more time playing an online game or navigating a webpage, the person converting by buying a product or service, and so on. The reinforced learning can be based on exploring a solution space and adapting the model, algorithm, etc., based on outcomes of the exploration. When positive outcomes are encountered, the positive outcomes can be reinforced by changing weighting values within the model, algorithm, etc. Positive outcomes may result in increasing weighting values. Negative outcomes can also be considered, where weighting values may be reduced or otherwise adjusted.

FIG. 9 is a system diagram for an interior of a vehicle 900. Vehicle manipulation can use convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The evaluating provides a cognitive state analysis. The cognitive state analysis can be used for vehicle manipulation, for example. One or more occupants of a vehicle 910, such as occupants 920 and 922, can be observed using a microphone 940, one or more cameras 942, 944, or 946, and other audio and image capture techniques. The image data can include video data. The video data and the audio data can include cognitive state data, where the cognitive state data can include facial data, voice data, physiological data, and the like. The occupant can be a driver occupant 922 of the vehicle 910, a passenger occupant 920 within the vehicle, and so on.

The cameras or imaging devices that can be used to obtain images including facial data from the occupants of the vehicle 910 can be positioned to capture the face of the vehicle operator, the face of a vehicle passenger, multiple views of the faces of occupants of the vehicle, and so on. The cameras can be located near a rear-view mirror 914, such as camera 942, positioned near or on a dashboard 916, such as camera 944, positioned within the dashboard, such as camera 946, and so on. The microphone, or audio capture device, 940 can be positioned within the vehicle such that voice data, speech data, non-speech vocalizations, and so on, can be easily collected with minimal background noise. In embodiments, additional cameras, imaging devices, microphones, audio capture devices, and so on, can be located throughout the vehicle. In further embodiments, each occupant of the vehicle could have multiple cameras, microphones, etc., positioned to capture video data and audio data from that occupant.

The interior of a vehicle 910 can be a standard vehicle, an autonomous vehicle, a semi-autonomous vehicle, and so on. The vehicle can be a sedan or other automobile, a van, a sport utility vehicle (SUV), a truck, a bus, a special purpose vehicle, and the like. The interior of the vehicle 910 can include standard controls such as a steering wheel 936, a throttle control (not shown), a brake 934, and so on. The interior of the vehicle can include other controls 932 such as controls for seats, mirrors, climate adjustments, audio systems, etc. The controls 932 of the vehicle 910 can be controlled by a controller 930. The controller 930 can control the vehicle 910 in various manners such as autonomously, semi-autonomously, assertively to a vehicle occupant 920 or 922, etc. In embodiments, the controller provides vehicle control techniques, assistance, etc. The controller 930 can receive instructions via an antenna 912 or using other wireless techniques. The controller 930 can be preprogrammed to cause the vehicle to follow a specific route. The specific route that the vehicle is programmed to follow can be based on the cognitive state of the vehicle occupant. The specific route can be chosen based on lowest stress, least traffic, best view, shortest route, and so on.

FIG. 10 is a timeline with information tracks relating to cognitive states. A timeline can show one or more cognitive states that can be experienced, for example, by a vehicle occupant. The timeline can be based on vehicular cognitive data collection using multiple devices. A computer is initialized for convolutional processing. Using an imaging device, images of, for example, an occupant of a first vehicle are obtained. A multilayered analysis engine is trained on the computer using the images. Further images, including facial image data from persons present in a second vehicle, are evaluated, and manipulation data is provided to the second vehicle. The timeline 1010 with information tracks 1000 relates to various cognitive states. A first track 1060 shows events that, in embodiments, are related to use of a computer by the individual. A first event 1020 can indicate an action that the individual took (such as launching an application); an action initiated by the computer (such as the presentation of a dialog box); an external event (such as a new global positioning system (GPS) coordinate); or another event such as receiving an email, a phone call, a text message, or any other type of event. In some embodiments, a photograph can be used to document an event or simply to save contextual information in the first track 1060. A second event 1022 can indicate another action or event in a similar manner. Such events can be used to provide contextual information and can also include information such as copies of emails, text messages, phone logs, file names, or other information that can prove useful in understanding the context of a user's actions. Thus, in embodiments, contextual information is based on one or more of a photograph, an email, a text message, a phone log, or GPS information.

A second track 1062 can include continuously collected cognitive state data 1030, such as electrodermal activity data. A third track 1064 can include facial data. The facial data can be collected intermittently when the individual is looking toward a camera. The facial data 1040 can include one or more still photographs, videos, or abstracted facial expressions which can be collected when the user looks in the direction of the camera. A fourth track 1066 can include facial data that is collected either intermittently or continuously by a second camera. The facial data 1042 can include one or more still photographs, videos, or abstracted facial expressions which can be collected when the user looks in the direction of that camera. A fifth track 1068 can include facial data that is collected from a third camera, such as the webcam. In the example shown, the fifth track 1068 includes first facial data 1044, second facial data 1046, and third facial data 1048, which can be any type of facial data including data that can be used for determining cognitive state information. Any number of samples of facial data can be collected in any track. The cognitive state data from the various tracks can be collected simultaneously, collected on one track exclusive of other tracks, collected where cognitive state data overlaps between the tracks, and so on. When cognitive state data from multiple tracks overlap, one track's data can take precedence or the data from the multiple tracks can be combined.

Additional tracks, through the nth track 1070, of cognitive state data of any type can be collected. The additional tracks 1070 can be collected on a continuous or on an intermittent basis. The intermittent basis can be either occasional or periodic. Analysis can further comprise interpolating cognitive state data when the cognitive state data collected is intermittent, and/or imputing additional cognitive state data where the cognitive state data is missing. One or more interpolated tracks 1072 can be included and can be associated with cognitive state data that is collected on an intermittent basis, such as the facial data of the fifth track 1068. Interpolated data 1050 and further interpolated data 1052 can contain interpolations of the facial data of the fifth track 1068 for the time periods where no facial data was collected in that track. Other embodiments interpolate data for periods where no track includes facial data. In other embodiments, analysis includes interpolating cognitive state analysis when the cognitive state data collected is intermittent.

The cognitive state data, such as the continuous cognitive state data 1030 and/or any of the collected facial data 1040, 1042, 1044, 1046, and 1048, can be tagged. The tags can include metadata related to the cognitive state data, including, but not limited to, the device that collected the cognitive state data; the individual from whom the cognitive state data was collected; the task being performed by the individual; the media being viewed by the individual; and the location, environcognitive conditions, time, date, or any other contextual information. The tags can be used to locate pertinent cognitive state data; for example, the tags can be used to retrieve the cognitive state data from a database. The tags can be included with the cognitive state data that is sent over the Internet to cloud or web-based storage and/or services. As such, the tags can be used locally on the machine where the cognitive state data was collected and/or remotely on a remote server or a cloud/web service.

Other tags can be related to the cognitive state data. Further embodiments can include tagging the cognitive state data with sensor data. The sensor data can be obtained from the vehicle occupant along with the obtaining of the video data or the audio data, instead of obtaining the video data or the audio data, etc. In embodiments, the sensor data can include one or more of vehicle temperature, outside temperature, time of day, amount of daylight, weather conditions, headlight activation, windshield wiper activation, entertainment center selection, or entertainment center volume. Other sensor data can include physiological data related to one or more occupants of the vehicle. The physiological data can include heart rate, heart rate variability, electrodermal activity, acceleration, and the like. The tags can also be related to the cognitive state that can be determined by image-based analysis of the video, audio, or physiological data, or other techniques. In embodiments, the tags that can be applied can be based on one or more of drowsiness, fatigue, distraction, impairment, sadness, stress, happiness, anger, frustration, confusion, disappointment, hesitation, cognitive overload, focusing, engagement, attention, boredom, exploration, confidence, trust, delight, disgust, skepticism, doubt, satisfaction, excitement, laughter, calmness, curiosity, humor, depression, envy, sympathy, embarrassment, poignancy, or mirth.

FIG. 11 shows cognitive state data with tags. Cognitive state data, including cognitive state data with tags, can be used in convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The cognitive state data, such as the cognitive state data with tags 1100, includes video image cognitive state data 1110 captured on an individual from a first source. In some embodiments, the source of the cognitive state data includes certain standard metadata 1120 with the cognitive state data 1110. For example, a video camera which includes timestamps along with video data demonstrates such metadata inclusion. A still camera which includes EXIF (or Exif) data identifying the camera model, exposure information, and day and date information in the JPEG or other image file format containing the compressed image data shows another instance of metadata inclusion.

In embodiments, additional data which provides information about the cognitive state data 1110 is determined. Such additional data can be tagged to the cognitive state data as cognitive state metadata 1130. The cognitive state metadata 1130 can provide information about the cognitive states useful in the analysis of the cognitive state data 1110. In embodiments, the cognitive state can include one or more of drowsiness, fatigue, distraction, impairment, sadness, stress, happiness, anger, frustration, confusion, disappointment, hesitation, cognitive overload, focusing, engagement, attention, boredom, exploration, confidence, trust, delight, disgust, skepticism, doubt, satisfaction, excitement, laughter, calmness, curiosity, humor, depression, envy, sympathy, embarrassment, poignancy, or mirth. The cognitive state metadata 1130, or additional data, is data that is not tagged to the cognitive state data by the source of the cognitive state data and not always known to the source of the cognitive state data 1110. Thus, the cognitive state metadata 1130 is tagged to the cognitive state data 1110 by an entity that is not the original source of the cognitive state data.

In one embodiment, a video camera is used to capture the cognitive state data 1110. The video camera can include standard metadata 1120 such as time and date and model number of the camera, along with the video image, which in this case comprises video image cognitive state data 1110, in a MPEG-4 data stream that is sent from the video camera to a cognitive state data collection machine. The standard metadata 1120 can be included using standard metadata formats defined by the MPEG-4 specification. The cognitive state data collection machine can determine an identity of the individual being monitored, based on a login ID, and an activity of that individual, such as watching a particular media presentation. The cognitive state data collection machine can then tag the video image with the login ID and the name of the particular media presentation as cognitive state metadata 1130. In at least one embodiment, the cognitive state data collection machine formats the cognitive state metadata as XMP metadata and includes it in the MPEG-4 file. Other embodiments determine different additional information to be used as cognitive state metadata 1130 and use different formats to tag the cognitive state data 1110 with the cognitive state metadata 1130.

Once the data collection machine has captured cognitive state data, at least a portion of the cognitive state data tagged with the additional data is sent to a web service. The portion of the cognitive state data sent to the web service can be based on the additional contextual data collected or can be based on cognitive state metadata 1130. At the web service, portions of cognitive state data can be selected for analysis which is based, at least in part, on tags identifying one or more contexts. In at least one embodiment, the selected portions are based, at least in part, on identifying a particular individual. In some embodiments, the selected portions include tags identifying at least two different timestamps so that samples can be distributed over a period of time. In some embodiments, the selected portions are based, at least in part, on tags identifying a particular context. Once the portions are selected, they can be analyzed by the web service and can be used to create cognitive state information.

FIG. 12 shows example image and audio collection including multiple mobile devices. Cognitive state data including image data, facial data, audio data, voice data, physiological data, and so on, can be collected using multiple mobile devices. The collected cognitive state data can be produced by a convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. While one person is shown, in practice the video data or audio data on any number of people can be collected. In the diagram 1200, the multiple mobile devices can be used separately or in any combination to collect video data, audio data, or both video and audio data on a user 1210. While one person is shown, the video data and audio data can be collected on multiple people. A user 1210 can be observed as she or he is performing a task, experiencing an event, viewing a media presentation, and so on. The user 1210 can be shown one or more media presentations, political presentations, social media, or another form of displayed media. The one or more media presentations can be shown to a plurality of people. The media presentations can be displayed on an electronic display 1212 or another display. The data collected on the user 1210 or on a plurality of users can be in the form of one or more videos, video frames, and still images; one or more audio channels, etc. The plurality of video data and audio data can be of people who are experiencing different situations. Some example situations can include the user or plurality of users being exposed to TV programs, movies, video clips, social media, and other such media. The situations could also include exposure to media such as advertisements, political messages, news programs, and so on.

As noted before, video data and audio data can be collected on one or more users in substantially identical or different situations while viewing either a single media presentation or a plurality of presentations. The data collected on the user 1210 can be analyzed and viewed for a variety of purposes including expression analysis, cognitive state analysis, mental state analysis, emotional state analysis, and so on. The electronic display 1212 can be on a laptop computer 1220 as shown, a tablet computer 1250, a cell phone 1240, a television, a mobile monitor, or any other type of electronic device. In one embodiment, video data including expression data is collected on a mobile device such as a cell phone 1240, a tablet computer 1250, a laptop computer 1220, or a watch 1270. Similarly, the audio data including speech data and non-speech vocalizations can be collected on one or more of the mobile devices. Thus, the multiple sources can include at least one mobile device, such as a phone or cell phone 1240 or a tablet 1250, or a wearable device such as a watch 1270 or glasses 1260. A mobile device can include a forward-facing camera and/or a rear-facing camera that can be used to collect expression data. A mobile device can include a microphone, audio transducer, or other audio capture apparatus that can be used to capture the speech and non-speech vocalizations. Sources of expression data can include a webcam 1222, a phone camera 1242, a tablet camera 1252, a wearable camera 1262, and a mobile camera 1230. A wearable camera can comprise various camera devices, such as a watch camera 1272. Sources of audio data 1282 can include a microphone 1280.

As the user 1210 is monitored, the user might move due to the nature of the task, boredom, discomfort, distractions, or for another reason. As the user moves, the camera with a view of the user's face can be changed. Thus, as an example, if the user is looking in a first direction, the line of sight 1224 from the webcam 1222 is able to observe the user's face, but if the user is looking in a second direction, the line of sight 1234 from the mobile camera 1230 is able to observe the user's face. Furthermore, in other embodiments, if the user is looking in a third direction, the line of sight 1244 from the phone camera 1242 is able to observe the user's face, and if the user is looking in a fourth direction, the line of sight 1254 from the tablet camera 1252 is able to observe the user's face. If the user is looking in a fifth direction, the line of sight 1264 from the wearable camera 1262, which can be a device such as the glasses 1260 shown and can be worn by another user or an observer, is able to observe the user's face. If the user is looking in a sixth direction, the line of sight 1274 from the wearable watch-type device 1270, with a camera 1272 included on the device, is able to observe the user's face. In other embodiments, the wearable device is another device, such as an earpiece with a camera, a helmet or hat with a camera, a clip-on camera attached to clothing, or any other type of wearable device with a camera or other sensor for collecting expression data. The user 1210 can also use a wearable device including a camera for gathering contextual information and/or collecting expression data on other users. Because the user 1210 can move her or his head, the facial data can be collected intermittently when she or he is looking in a direction of a camera. In some cases, multiple people can be included in the view from one or more cameras, and some embodiments include filtering out faces of one or more other people to determine whether the user 1210 is looking toward a camera. All or some of the expression data can be continuously or sporadically available from the various devices and other devices.

The captured video data can include facial expressions and can be analyzed on a computing device such as the video capture device, or on another separate device. The analysis can take place on one of the mobile devices discussed above, on a local server, on a remote server, and so on. In embodiments, some of the analysis takes place on the mobile device, while other analysis takes place on a server device. The analysis of the video data can include the use of a classifier. The video data can be captured using one of the mobile devices discussed above and can be sent to a server or another computing device for analysis. However, the captured video data including expressions can also be analyzed on the device which performed the capturing. The analysis can be performed on a mobile device, where the videos were obtained with the mobile device and wherein the mobile device includes one or more of a laptop computer, a tablet, a PDA, a smartphone, a wearable device, and so on. In another embodiment, the analyzing comprises using a classifier on a server or another computing device other than the capturing device.

FIG. 13 illustrates feature extraction for multiple faces. Image analysis and image processing, including facial analysis and processing, can be based on feature extraction from multiple faces, which can be applied in convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The feature extraction for multiple faces can be performed for faces that can be detected in multiple images. In embodiments, the features of multiple faces are extracted for evaluating cognitive states. Features of a face or a plurality of faces can be extracted from collected video data. The feature extraction can be performed by analysis, by using one or more processors, by using one or more video collection devices, and by using a server. The analysis device can be used to perform face detection for a second face, as well as to perform facial tracking of the first face. One or more videos can be captured, where the videos contain one or more faces. The video or videos that contain the one or more faces can be partitioned into a plurality of frames, and the frames can be analyzed for the detection of the one or more faces. The analysis of the one or more video frames can be based on one or more classifiers. A classifier can be an algorithm, heuristic, function, or piece of code that can be used to identify into which of a set of categories a new or existing observation, sample, datum, etc. should be placed. The decision to place an observation into a category can be based on training the algorithm or piece of code by analyzing a known set of data, known as a training set. The training set can include data for which category memberships of the data can be known. The training set can be used as part of a supervised training technique. If a training set is not available, then a clustering technique can be used to group observations into categories. The latter approach, or unsupervised learning, can be based on a measure (i.e. distance) of one or more inherent similarities among the data that is being categorized. When a new observation is received, then the classifier can be used to categorize the new observation. Classifiers can be used for many analysis applications, including analysis of one or more faces. The use of classifiers can be the basis of analyzing the one or more faces for gender, ethnicity, and age; of detecting the one or more faces in one or more videos; of detecting facial features and landmarks; and so on. The observations can be analyzed based on one or more of a set of quantifiable properties. The properties can be described as features and explanatory variables involving various data types that can include numerical (integer-valued, real-valued), ordinal, categorical, and so on. Some classifiers can be based on a comparison between an observation and prior observations, as well as based on functions such as a similarity function, a distance function, and so on.

Classification can be based on various types of algorithms, heuristics, codes, procedures, statistics, and so on. Many techniques for performing classification exist. This classification of one or more observations into one or more groups can be based on distributions of the data values, probabilities, and so on. Classifiers can be binary, multiclass, linear, etc. Algorithms for classification can be implemented using a variety of techniques, including neural networks, kernel estimation, support vector machines, use of quadratic surfaces, and so on. Classification can be used in many application areas such as computer vision and speech and handwriting recognition. Classification can be used for biometric identification of one or more people in a single frame or in multiple frames of one or more videos.

Returning to FIG. 13, the detection of the first face, the second face, and multiple faces can include identifying facial landmarks, generating a bounding box, and predicting of a bounding box and landmarks for a next frame, where the next frame can be one of a plurality of frames of a video containing faces. A first video frame 1300 includes a frame boundary 1310, a first face 1312, and a second face 1314. The video frame 1300 also includes a bounding box 1320. Facial landmarks can be generated for the first face 1312. Face detection can be performed to initialize a second set of locations for a second set of facial landmarks for a second face within the video. Facial landmarks in the video frame 1300 can include the facial landmarks 1322, 1324, and 1326. The facial landmarks can include corners of a mouth, corners of eyes, eyebrow corners, the tip of the nose, nostrils, chin, the tips of ears, and so on. The performing of face detection on the second face can include performing facial landmark detection with the first frame from the video for the second face and can include estimating a second rough bounding box for the second face based on the facial landmark detection. The estimating of a second rough bounding box can include the bounding box 1320. Bounding boxes can also be estimated for one or more other faces within the boundary 1310. The bounding box can be refined, as can one or more facial landmarks. The refining of the second set of locations for the second set of facial landmarks can be based on localized information around the second set of facial landmarks. The bounding box 1320 and the facial landmarks 1322, 1324, and 1326 can be used to estimate future locations for the second set of locations for the second set of facial landmarks in a future video frame from the first video frame.

A second video frame 1302 is also shown. The second video frame 1302 includes a frame boundary 1330, a first face 1332, and a second face 1334. The second video frame 1302 also includes a bounding box 1340 and the facial landmarks, or points, 1342, 1344, and 1346. In other embodiments, multiple facial landmarks are generated and used for facial tracking of the two or more faces of a video frame, such as the shown second video frame 1302. Facial points from the first face can be distinguished from other facial points. In embodiments, the other facial points include facial points of one or more other faces. The facial points can correspond to the facial points of the second face. The distinguishing of the facial points of the first face and the facial points of the second face can be used to differentiate between the first face and the second face, to track either the first face, the second face, or both faces, and so on. Other facial points can correspond to the second face. As mentioned above, multiple facial points can be determined within a frame. One or more of the other facial points that are determined can correspond to a third face. The location of the bounding box 1340 can be estimated, where the estimating can be based on the location of the generated bounding box 1320 shown in the first video frame 1300. The three facial points shown, facial points, or landmarks, 1342, 1344, and 1346, might lie within the bounding box 1340 or might not lie partially or completely within the bounding box 1340. For instance, the second face 1334 might have moved between the first video frame 1300 and the second video frame 1302. Based on the accuracy of the estimating of the bounding box 1340, a new estimation can be determined for a third, future frame from the video, and so on. The evaluation can be performed, all or in part, using semiconductor-based logic.

FIG. 14 shows an example of livestreaming of social video and audio. The streaming of social video and social audio can be applied to convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The livestreaming and image analysis can be facilitated by a video capture device, a local server, a remote server, a semiconductor-based logic, and so on. The streaming can be livestreaming and can include cognitive state analysis, cognitive state event signature analysis, etc. Livestreaming video is an example of one-to-many social media, where video can be sent over the Internet from one person to a plurality of people using a social media app and/or platform. Livestreaming is one of numerous popular techniques used by people who want to disseminate ideas, send information, provide entertainment, share experiences, and so on. Some of the livestreams can be scheduled, such as webcasts, online classes, sporting events, news, computer gaming, or video conferences, while others can be impromptu streams that are broadcast as needed or when desirable. Examples of impromptu livestream videos can range from individuals simply wanting to share experiences with their social media followers, to live coverage of breaking news, emergencies, or natural disasters. The latter coverage is known as mobile journalism, or “mo jo”, and is becoming increasingly common. With this type of coverage, news reporters can use networked, portable electronic devices to provide mobile journalism content to a plurality of social media followers. Such reporters can be quickly and inexpensively deployed as the need or desire arises.

Several livestreaming social media apps and platforms can be used for transmitting video. One such video social media app is Meerkat™ which can link with a user's Twitter™ account. Meerkat™ enables a user to stream video using a handheld, networked electronic device coupled to video capabilities. Viewers of the livestream can comment on the stream using tweets that can be seen and responded to by the broadcaster. Another popular app is Periscope™ which can transmit a live recording from one user to his or her Periscope™ account and to other followers. The Periscope™ app can be executed on a mobile device. The user's Periscope™ followers can receive an alert whenever that user begins a video transmission. Another livestream video platform is Twitch™ which can be used for video streaming of video gaming and broadcasts of various competitions and events.

The example 1400 shows a user 1410 broadcasting a video livestream and an audio livestream to one or more people as shown by a first person 1450, a second person 1460, and a third person 1470. A portable, network-enabled, electronic device 1420 can be coupled to a front-side camera 1422. The portable electronic device 1420 can be a smartphone, a PDA, a tablet, a laptop computer, and so on. The front-side camera 1422 coupled to the device 1420 can have a line-of-sight view 1424 to the user 1410 and can capture video of the user 1410. The portable electronic device 1420 can be coupled to a microphone (not shown). The microphone can capture voice data 1428 such as speech and non-speech vocalizations. In embodiments, non-speech vocalizations can include grunts, yelps, squeals, snoring, sighs, laughter, filled pauses, unfilled pauses, yawns, or the like. The captured video and audio can be sent to an analysis machine or recommendation engine 1440 using a network link 1426 to the Internet 1430. The network link can be a wireless link, a wired link, and so on. The recommendation engine 1440 can recommend to the user 1410 an app and/or platform that can be supported by the server and can be used to provide a video livestream, an audio livestream, or both a video livestream and an audio livestream to one or more followers of the user 1410.

In the example 1400, the user 1410 has three followers: a first person 1450, a second person 1460, and a third person 1470. Each follower has a line-of-sight view to a video screen on a portable, networked electronic device. In other embodiments, one or more followers follow the user 1410 using any other networked electronic device, including a computer. In the example 1400, a first person 1450 has a line-of-sight view 1452 to the video screen of a device 1454; a second person 1460 has a line-of-sight view 1462 to the video screen of a device 1464, and a third person 1470 has a line-of-sight view 1472 to the video screen of a device 1474. The device 1474 can also capture audio data 1478 from the third person 1470. The portable electronic devices 1454, 1464, and 1474 can each be a smartphone, a PDA, a tablet, and so on. Each portable device can receive the video stream and the audio stream being broadcast by the user 1410 through the Internet 1430 using the app and/or platform that can be recommended by the recommendation engine 1440. The device 1454 can receive a video stream and the audio stream using the network link 1456, the device 1464 can receive a video stream and the audio stream using the network link 1466, the device 1474 can receive a video stream and the audio stream using the network link 1476, and so on. The network link can be a wireless link, a wired link, a hybrid link, and so on. Depending on the app and/or platform that can be recommended by the recommendation engine 1440, one or more followers, such as the followers 1450, 1460, and 1470 shown, can reply to, comment on, or otherwise provide feedback to the user 1410 using their respective devices 1454, 1464, and 1474.

The human face provides a powerful communications medium through its ability to exhibit numerous expressions that can be captured and analyzed for a variety of purposes. In some cases, media producers are acutely interested in evaluating the effectiveness of message delivery by video media. Such video media includes advertisements, political messages, educational materials, television programs, movies, government service announcements, etc. Automated facial analysis can be performed on one or more video frames containing a face in order to detect facial action. Based on the facial action detected, a variety of parameters can be determined, including affect valence, spontaneous reactions, facial action units, and so on. The parameters that are determined can be used to infer or predict emotional, mental, and cognitive states. For example, determined valence can be used to describe the emotional reaction of a viewer to a video media presentation or another type of presentation. Positive affect valence provides evidence that a viewer is experiencing a favorable emotional response to the video media presentation, while negative affect valence provides evidence that a viewer is experiencing an unfavorable emotional response to the video media presentation. Other facial data analysis can include the determination of discrete emotional states of the viewer or viewers.

Facial data can be collected from a plurality of people using any of a variety of cameras. A camera can include a webcam, a video camera, a still camera, a thermal imager, a CCD device, a phone camera, a three-dimensional camera, a depth camera, a light field camera, multiple webcams used to show different views of a person, or any other type of image capture apparatus that can allow captured data to be used in an electronic system. In some embodiments, the person is permitted to “opt in” to the facial data collection. For example, the person can agree to the capture of facial data using a personal device such as a mobile device or another electronic device by selecting an opt-in choice. Opting-in can then turn on the person's webcam-enabled device and can begin the capture of the person's facial data via a video feed from the webcam or other camera. The video data that is collected can include one or more persons experiencing an event. The one or more persons can be sharing a personal electronic device or can each be using one or more devices for video capture. The videos that are collected can be collected using a web-based framework. The web-based framework can be used to display the video media presentation or event as well as to collect videos from multiple viewers who are online. That is, the collection of videos can be crowdsourced from those viewers who elected to opt in to the video data collection.

The videos captured from the various viewers who chose to opt in can be substantially different in terms of video quality, frame rate, etc. As a result, the facial video data can be scaled, rotated, and otherwise adjusted to improve consistency. Human factors further contribute to the capture of the facial video data. The facial data that is captured might or might not be relevant to the video media presentation being displayed. For example, the viewer might not be paying attention, might be fidgeting, might be distracted by an object or event near the viewer, or might be otherwise inattentive to the video media presentation. The behavior exhibited by the viewer can prove challenging to analyze due to viewer actions including eating, speaking to another person or persons, speaking on the phone, etc. The videos collected from the viewers might also include other artifacts that pose challenges during the analysis of the video data. The artifacts can include items such as eyeglasses (because of reflections), eye patches, jewelry, and clothing that occludes or obscures the viewer's face. Similarly, a viewer's hair or hair covering can present artifacts by obscuring the viewer's eyes and/or face.

The captured facial data can be analyzed using the facial action coding system (FACS). The FACS seeks to define groups or taxonomies of facial movements of the human face. The FACS encodes movements of individual muscles of the face, where the muscle movements often include slight, instantaneous changes in facial appearance. The FACS encoding is commonly performed by trained observers, but can also be performed on automated, computer-based systems. Analysis of the FACS encoding can be used to determine emotions of the persons whose facial data is captured in the videos. The FACS is used to encode a wide range of facial expressions that are anatomically possible for the human face. The FACS encodings include action units (AUs) and related temporal segments that are based on the captured facial expression. The AUs are open to higher order interpretation and decision-making. These AUs can be used to recognize emotions experienced by the person who is being observed. Emotion-related facial actions can be identified using the emotional facial action coding system (EMFACS) and the facial action coding system affect interpretation dictionary (FACSAID). For a given emotion, specific action units can be related to the emotion. For example, the emotion of anger can be related to AUs 4, 5, 7, and 23, while happiness can be related to AUs 6 and 12. Other mappings of emotions to AUs have also been previously associated. The coding of the AUs can include an intensity scoring that ranges from A (trace) to E (maximum). The AUs can be used for analyzing images to identify patterns indicative of a particular cognitive and/or emotional state. The AUs range in number from 0 (neutral face) to 98 (fast up-down look). The AUs include so-called main codes (inner brow raiser, lid tightener, etc.), head movement codes (head turn left, head up, etc.), eye movement codes (eyes turned left, eyes up, etc.), visibility codes (eyes not visible, entire face not visible, etc.), and gross behavior codes (sniff, swallow, etc.). Emotion scoring can be included where intensity is evaluated, and specific emotions, moods, mental states, or cognitive states can be identified.

The coding of faces identified in videos captured of people observing an event can be automated. The automated systems can detect facial AUs or discrete emotional states. The emotional states can include amusement, fear, anger, disgust, surprise, and sadness. The automated systems can be based on a probability estimate from one or more classifiers, where the probabilities can correlate with an intensity of an AU or an expression. The classifiers can be used to identify into which of a set of categories a given observation can be placed. In some cases, the classifiers can be used to determine a probability that a given AU or expression is present in a given frame of a video. The classifiers can be used as part of a supervised machine learning technique, where the machine learning technique can be trained using “known good” data. Once trained, the machine learning technique can proceed to classify new data that is captured.

The supervised machine learning models can be based on support vector machines (SVMs). An SVM can have an associated learning model that is used for data analysis and pattern analysis. For example, an SVM can be used to classify data that can be obtained from collected videos of people experiencing a media presentation. An SVM can be trained using “known good” data that is labeled as belonging to one of two categories (e.g. smile and no-smile). The SVM can build a model that assigns new data into one of the two categories. The SVM can construct one or more hyperplanes that can be used for classification. The hyperplane that has the largest distance from the nearest training point can be determined to have the best separation. The largest separation can improve the classification technique by increasing the probability that a given data point can be properly classified.

In another example, a histogram of oriented gradients (HoG) can be computed. The HoG can include feature descriptors and can be computed for one or more facial regions of interest. The regions of interest of the face can be located using facial landmark points, where the facial landmark points can include outer edges of nostrils, outer edges of the mouth, outer edges of eyes, etc. A HoG for a given region of interest can count occurrences of gradient orientation within a given section of a frame from a video, for example. The gradients can be intensity gradients and can be used to describe an appearance and a shape of a local object. The HoG descriptors can be determined by dividing an image into small, connected regions, also called cells. A histogram of gradient directions or edge orientations can be computed for pixels in the cell. Histograms can be contrast-normalized based on intensity across a portion of the image or the entire image, thus reducing any influence from differences in illumination or shadowing changes between and among video frames. The HoG can be computed on the image or on an adjusted version of the image, where the adjustment of the image can include scaling, rotation, etc. The image can be adjusted by flipping the image around a vertical line through the middle of a face in the image. The symmetry plane of the image can be determined from the tracker points and landmarks of the image.

In embodiments, an automated facial analysis system identifies five facial actions or action combinations in order to detect spontaneous facial expressions for media research purposes. Based on the facial expressions that are detected, a determination can be made with regard to the effectiveness of a given video media presentation, for example. The system can detect the presence of the AUs or the combination of AUs in videos collected from a plurality of people. The facial analysis technique can be trained using a web-based framework to crowdsource videos of people as they watch online video content. The video can be streamed at a fixed frame rate to a server. Human labelers can code for the presence or absence of facial actions including a symmetric smile, unilateral smile, asymmetric smile, and so on. The trained system can then be used to automatically code the facial data collected from a plurality of viewers experiencing video presentations (e.g. television programs).

Spontaneous asymmetric smiles can be detected in order to understand viewer experiences. Related literature indicates that as many asymmetric smiles occur on the right hemi face as do on the left hemi face for spontaneous expressions. Detection can be treated as a binary classification problem, where images that contain a right asymmetric expression are used as positive (target class) samples and all other images as negative (non-target class) samples. Classifiers perform the classification, including classifiers such as support vector machines (SVMs) and random forests. Random forests can include ensemble-learning methods that use multiple learning algorithms to obtain better predictive performance. Frame-by-frame detection can be performed to recognize the presence of an asymmetric expression in each frame of a video. Facial points can be detected, including the top of the mouth and the two outer eye corners. The face can be extracted, cropped, and warped into a pixel image of specific dimension (e.g. 96×96 pixels). In embodiments, the inter-ocular distance and vertical scale in the pixel image are fixed. Feature extraction can be performed using computer vision software such as OpenCV™. Feature extraction can be based on the use of HoGs. HoGs can include feature descriptors and can be used to count occurrences of gradient orientation in localized portions or regions of the image. Other techniques can be used for counting occurrences of gradient orientation, including edge orientation histograms, scale-invariant feature transformation descriptors, etc. The AU recognition tasks can also be performed using Local Binary Patterns (LBPs) and Local Gabor Binary Patterns (LGBPs). The HoG descriptor represents the face as a distribution of intensity gradients and edge directions and is robust in its ability to translate and scale. Differing patterns, including groupings of cells of various sizes and arranged in variously sized cell blocks, can be used. For example, 4×4 cell blocks of 8×8-pixel cells with an overlap of half of the block can be used. Histograms of channels can be used, including nine channels or bins evenly spread over 0-180 degrees. In this example, the HoG descriptor on a 96×96 image is 25 blocks×16 cells×9 bins=3600, the latter quantity representing the dimension. AU occurrences can be rendered. The videos can be grouped into demographic datasets based on nationality and/or other demographic parameters for further detailed analysis. This grouping and other analyses can be facilitated via semiconductor-based logic.

FIG. 15 is a flow diagram for detecting facial expressions. Cognitive states can be determined by detecting and analyzing facial expressions in images. The cognitive states can be evaluated by convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The flow 1500, or portions thereof, can be implemented in semiconductor logic, can be accomplished using a mobile device, can be accomplished using a server device, and so on. The flow 1500 can be used to automatically detect a wide range of facial expressions. A facial expression can produce strong emotional signals that can indicate valence and discrete emotional states. The discrete emotional states can include contempt, doubt, defiance, happiness, fear, anxiety, and so on. The detection of facial expressions can be based on the location of facial landmarks. The detection of facial expressions can be based on determination of action units (AUs), where the action units are determined using FACS coding. The AUs can be used separately or in combination to identify facial expressions. Based on the facial landmarks, one or more AUs can be identified by number and intensity. For example, AU12 can be used to code a lip corner puller and can be used to infer a smirk.

The flow 1500 begins by obtaining training image samples 1510. The image samples can include a plurality of images of one or more people. Human coders who are trained to correctly identify AU codes based on the FACS can code the images. The training or “known good” images can be used as a basis for training a machine learning technique. Once trained, the machine learning technique can be used to identify AUs in other images that can be collected using a camera, a sensor, and so on. The flow 1500 continues with receiving an image 1520. The image 1520 can be received from a camera, a sensor, and so on. As previously discussed, the camera or cameras can include a webcam, where a webcam can include a video camera, a still camera, a thermal imager, a CCD device, a phone camera, a three-dimensional camera, a depth camera, a light field camera, multiple webcams used to show different views of a person, or any other type of image capture apparatus that can allow captured data to be used in an electronic system. The image that is received can be manipulated in order to improve the processing of the image. For example, the image can be cropped, scaled, stretched, rotated, flipped, etc. in order to obtain a resulting image that can be analyzed more efficiently. Multiple versions of the same image can be analyzed. In some cases, the manipulated image and a flipped or mirrored version of the manipulated image can be analyzed alone and/or in combination to improve analysis. The flow 1500 continues with generating histograms 1530 for the training images and the one or more versions of the received image. The histograms can be based on a HoG or another histogram. As described in previous paragraphs, the HoG can include feature descriptors and can be computed for one or more regions of interest in the training images and the one or more received images. The regions of interest in the images can be located using facial landmark points, where the facial landmark points can include outer edges of nostrils, outer edges of the mouth, outer edges of eyes, etc. A HoG for a given region of interest can count occurrences of gradient orientation within a given section of a frame from a video.

The flow 1500 continues with applying classifiers 1540 to the histograms. The classifiers can be used to estimate probabilities, where the probabilities can correlate with an intensity of an AU or an expression. In some embodiments, the choice of classifiers used is based on the training of a supervised learning technique to identify facial expressions. The classifiers can be used to identify into which of a set of categories a given observation can be placed. The classifiers can be used to determine a probability that a given AU or expression is present in a given image or frame of a video. In various embodiments, the one or more AUs that are present include AU01 inner brow raiser, AU12 lip corner puller, AU38 nostril dilator, and so on. In practice, the presence or absence of multiple AUs can be determined. The flow 1500 continues with computing a frame score 1550. The score computed for an image, where the image can be a frame from a video, can be used to determine the presence of a facial expression in the image or video frame. The score can be based on one or more versions of the image 1520 or a manipulated image. The score can be based on a comparison of the manipulated image to a flipped or mirrored version of the manipulated image. The score can be used to predict a likelihood that one or more facial expressions are present in the image. The likelihood can be based on computing a difference between the outputs of a classifier used on the manipulated image and on the flipped or mirrored image, for example. The classifier that is used can identify symmetrical facial expressions (e.g. smile), asymmetrical facial expressions (e.g. outer brow raiser), and so on.

The flow 1500 continues with plotting results 1560. The results that are plotted can include one or more scores for one or more frames computed over a given time t. For example, the plotted results can include classifier probability results from analysis of HoGs for a sequence of images and video frames. The plotted results can be matched with a template 1562. The template can be temporal and can be represented by a centered box function or another function. A best fit with one or more templates can be found by computing a minimum error. Other best-fit techniques can include polynomial curve fitting, geometric curve fitting, and so on. The flow 1500 continues with applying a label 1570. The label can be used to indicate that a particular facial expression has been detected in the one or more images or video frames which constitute the image 1520 that was received. The label can be used to indicate that any of a range of facial expressions has been detected, including a smile, an asymmetric smile, a frown, and so on. Various steps in the flow 1500 may be changed in order, repeated, omitted, or the like without departing from the disclosed concepts. Various embodiments of the flow 1500 can be included in a computer program product embodied in a non-transitory computer readable medium that includes code executable by one or more processors. Various embodiments of the flow 1500, or portions thereof, can be included on a semiconductor chip and implemented in special purpose logic, programmable logic, and so on.

FIG. 16 is a flow diagram for the large-scale clustering of facial events. Large-scale clustering of facial events can be performed by convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. Cognitive state events can include facial events, speech events, etc. The large-scale clustering of facial events can be performed for data collected from a remote computing device. The facial events can be collected from people as they interact with a vehicle. The clustering and evaluation of facial events can be augmented using a mobile device, a server, semiconductor-based logic, and so on. As discussed above, collection of facial video data from one or more people can include a web-based framework. The web-based framework can be used to collect facial video data from large numbers of people located over a wide geographic area. The web-based framework can include an opt-in feature that allows people to agree to facial data collection. The web-based framework can be used to render and display data to one or more people and can collect data from the one or more people. For example, the facial data collection can be based on showing one or more viewers a video media presentation through a website. The web-based framework can be used to display the video media presentation or event and to collect videos from multiple viewers who are online. That is, the collection of videos can be crowdsourced from those viewers who elected to opt in to the video data collection. The video event can be a commercial, a political ad, an educational segment, and so on.

The flow 1600 includes obtaining videos containing faces 1610. The videos can be obtained using one or more cameras, where the cameras can include a webcam coupled to one or more devices employed by the one or more people using the web-based framework. The flow 1600 continues with extracting features from the individual responses 1620. The individual responses can include videos containing faces observed by the one or more webcams. The features that are extracted can include facial features such as an eyebrow, a nostril, an eye edge, a mouth edge, and so on. The feature extraction can be based on facial coding classifiers, where the facial coding classifiers output a probability that a specified facial action has been detected in a given video frame. The flow 1600 continues with performing unsupervised clustering of features 1630. The unsupervised clustering can be based on an event. The unsupervised clustering can be based on a K-Means, where the K of the K-Means can be computed using a Bayesian Information Criterion (BICk), for example, to determine the smallest value of K that meets system requirements. Any other criterion for K can be used. The K-Means clustering technique can be used to group one or more events into various respective categories.

The flow 1600 includes characterizing cluster profiles 1640. The profiles can include a variety of facial expressions such as smiles, asymmetric smiles, eyebrow raisers, eyebrow lowerers, etc. The profiles can be related to a given event. For example, a humorous video can be displayed in the web-based framework and the video data of people who have opted in can be collected. The characterization of the collected and analyzed video can depend in part on the number of smiles that occurred at various points throughout the humorous video. Similarly, the characterization can be performed on collected and analyzed videos of people viewing a news presentation. The characterized cluster profiles can be further analyzed based on demographic data. The number of smiles resulting from people viewing a humorous video can be compared across various demographic groups, where the groups can be formed based on geographic location, age, ethnicity, gender, and so on.

The flow 1600 can include determining cognitive state event temporal signatures 1650. The cognitive state event temporal signatures can include information on rise time to facial expression intensity, fall time from facial expression intensity, duration of a facial expression, and so on. In some embodiments, the cognitive state event temporal signatures are associated with certain demographics, ethnicities, cultures, etc. The cognitive state event temporal signatures can be used to identify one or more of sadness, stress, happiness, anger, frustration, confusion, disappointment, hesitation, cognitive overload, focusing, engagement, attention, boredom, exploration, confidence, trust, delight, disgust, skepticism, doubt, satisfaction, excitement, laughter, calmness, curiosity, humor, depression, envy, sympathy, embarrassment, poignancy, or mirth. Various steps in the flow 1600 may be changed in order, repeated, omitted, or the like without departing from the disclosed concepts. Various embodiments of the flow 1600 can be included in a computer program product embodied in a non-transitory computer readable medium that includes code executable by one or more processors. Various embodiments of the flow 1600, or portions thereof, can be included on a semiconductor chip and implemented in special purpose logic, programmable logic, and so on.

FIG. 17 shows unsupervised clustering of features and characterizations of cluster profiles. The features can be based on image data, audio data, physiological data, and so on, of an individual. Convolutional image processing can be performed by convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The clustering of features and characterizations of cluster profiles can be performed for data collected from a remote computing device. The clustering of features and characterizations of cluster profiles can be performed for people as they interact with other people, webpages, videogames, video streams, advertisements, and many other settings. The sub-sectional components can be used with performing the evaluation of content of the face. The sub-sectional components can be used to provide a context. Features including samples of facial data can be clustered using unsupervised clustering. Various clusters can be formed which include similar groupings of facial data observations. The example 1700 shows three clusters, clusters 1710, 1712, and 1714. The clusters can be based on video collected from people who have opted in to video collection. When the data collected is captured using a web-based framework, the data collection can be performed on a grand scale, including hundreds, thousands, or even more participants who can be located in close proximity and/or across a wide geographic area. Unsupervised clustering is a technique that can be used to process the large amounts of captured facial data and to identify groupings of similar observations. The unsupervised clustering can also be used to characterize the groups of similar observations. The characterizations can include identifying behaviors of the participants. The characterizations can be based on identifying facial expressions and facial action units of the participants. Some behaviors and facial expressions can include faster or slower onsets, faster or slower offsets, longer or shorter durations, etc. The onsets, offsets, and durations can all correlate to time. The data clustering that results from the unsupervised clustering can support data labeling. The labeling can include FACS coding. The clusters can be partially or totally based on a facial expression resulting from participants viewing a video presentation, where the video presentation can be an advertisement, a political message, educational material, a public service announcement, and so on. The clusters can be correlated with demographic information, where the demographic information can include educational level, geographic location, age, gender, income level, and so on.

The cluster profiles 1702 can be generated based on the clusters that can be formed from unsupervised clustering, with time shown on the x-axis and intensity or frequency shown on the y-axis. The cluster profiles can be based on captured facial data, including facial expressions. The cluster profile 1720 can be based on the cluster 1710, the cluster profile 1722 can be based on the cluster 1712, and the cluster profile 1724 can be based on the cluster 1714. The cluster profiles 1720, 1722, and 1724 can be based on smiles, smirks, frowns, or any other facial expression. The emotional states of the people who have opted in to video collection can be inferred by analyzing the clustered facial expression data. The cluster profiles can be plotted with respect to time and can show a rate of onset, a duration, and an offset (rate of decay). Other time-related factors can be included in the cluster profiles. The cluster profiles can be correlated with demographic information, as described above.

FIG. 18A shows example tags embedded in a webpage. Images can include facial images of a person or persons looking at a display, such as a webpage that is displayed on a mobile device. The images can be processed by convolutional computing using a multilayered analysis engine. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer. The tags embedded in the webpage can be used for image analysis and processing for data collected from a remote computing device. The tags embedded in the webpage can be used by people as they interact with a vehicle. Once a tag is detected, a mobile device, a server, semiconductor-based logic, etc. can be used to evaluate associated facial expressions. A webpage 1800 can include a page body 1810, a page banner 1812, and so on. The page body can include one or more objects, where the objects can include text, images, videos, audio, and so on. The example page body 1810 shown includes a first image, image 1 1820; a second image, image 2 1822; a first content field, content field 1 1840; and a second content field, content field 2 1842. In practice, the page body 1810 can contain multiple images and content fields and can include one or more videos, one or more audio presentations, and so on. The page body can include embedded tags, such as tag 1 1830 and tag 2 1832. In the example shown, tag 1 1830 is embedded in image 1 1820, and tag 2 1832 is embedded in image 2 1822. In embodiments, multiple tags are embedded. Tags can also be embedded in content fields, in videos, in audio presentations, etc. When a user mouses over a tag or clicks on an object associated with a tag, the tag can be invoked. For example, when the user mouses over tag 1 1830, tag 1 1830 can then be invoked. Invoking tag 1 1830 can include enabling a camera coupled to a user's device and capturing one or more images of the user as the user views a media presentation (or digital experience). In a similar manner, when the user mouses over tag 2 1832, tag 2 1832 can be invoked. Invoking tag 2 1832 can also include enabling the camera and capturing images of the user. In other embodiments, other actions are taken based on invocation of the one or more tags. Invoking an embedded tag can initiate an analysis technique, post to social media, award the user a coupon or another prize, initiate cognitive state analysis, perform emotion analysis, and so on.

FIG. 18B shows invoking tags to collect images. The invoking tags to collect images can be used for image analysis for data collected from a remote computing device. The invoking tags to collect images can be used for people as they interact with a vehicle. As previously stated, a media presentation can be a video, a webpage, and so on. A video 1802 can include one or more embedded tags, such as a tag 1860, a second tag 1862, a third tag 1864, a fourth tag 1866, and so on. In practice, multiple tags can be included in the media presentation. The one or more tags can be invoked during the media presentation. The collection of the invoked tags can occur over time, as represented by a timeline 1850. When a tag is encountered in the media presentation, the tag can be invoked. When the tag 1860 is encountered, invoking the tag can enable a camera coupled to a user device and can capture one or more images of the user viewing the media presentation. Invoking a tag can depend on opt-in by the user. For example, if a user has agreed to participate in a study by indicating an opt-in, then the camera coupled to the user's device can be enabled and one or more images of the user can be captured. If the user has not agreed to participate in the study and has indicated an opt-out, then invoking the tag 1860 neither enables the camera nor captures images of the user during the media presentation. The user can indicate an opt-in for certain types of participation, where opting-in can be dependent on specific content in the media presentation. For example, the user could opt in to participate in a study of political campaign messages and not opt in for a particular advertisement study. In this case, tags that are related to political campaign messages, advertising messages, social media sharing, etc. and that enable the camera and image capture when invoked would be embedded in the corresponding media presentation. However, tags embedded in the media presentation that are related to advertisements would not enable the camera when invoked. Various other situations of tag invocation are also possible.

FIG. 19 illustrates a bottleneck layer within a deep learning environment. A plurality of layers in a deep neural network (DNN) can include a bottleneck layer. The deep neural network can comprise a convolutional neural network. The bottleneck layer can be used in convolutional computing using a multilayered analysis engine. A deep neural network can apply classifiers such as image classifiers, audio classifiers, and so on. The classifiers can be learned by analyzing cognitive state data. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine includes multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer.

Layers of a deep neural network can include a bottleneck layer 1900. A bottleneck layer can be used for a variety of applications such as facial recognition, voice recognition, emotional state recognition, and so on. The deep neural network in which the bottleneck layer is located can include a plurality of layers. The plurality of layers can include an original feature layer 1910. A feature such as an image feature can include points, edges, objects, boundaries between and among regions, properties, and so on. The deep neural network can include one or more hidden layers 1920. The one or more hidden layers can include nodes, where the nodes can include nonlinear activation functions and other techniques. The bottleneck layer can be a layer that learns translation vectors to transform a neutral face to an emotional or expressive face. In some embodiments, the translation vectors can transform a neutral voice to an emotional or expressive voice. Specifically, activations of the bottleneck layer determine how the transformation occurs. A single bottleneck layer can be trained to transform a neutral face or voice to a different emotional face or voice. In some cases, an individual bottleneck layer can be trained for a transformation pair. At runtime, once the user's emotion has been identified and an appropriate response to it can be determined (mirrored or complementary), the trained bottleneck layer can be used to perform the needed transformation.

The deep neural network can include a bottleneck layer 1930. The bottleneck layer can include a fewer number of nodes than the one or more preceding hidden layers. The bottleneck layer can create a constriction in the deep neural network or other network. The bottleneck layer can force information that is pertinent to a classification, for example, into a low dimensional representation. The bottleneck features can be extracted using an unsupervised technique. In other embodiments, the bottleneck features can be extracted using a supervised technique. The supervised technique can include training the deep neural network with a known dataset. The features can be extracted from an autoencoder such as a variational autoencoder, a generative autoencoder, and so on. The deep neural network can include hidden layers 1940. The number of the hidden layers can include zero hidden layers, one hidden layer, a plurality of hidden layers, and so on. The hidden layers following the bottleneck layer can include more nodes than the bottleneck layer. The deep neural network can include a classification layer 1950. The classification layer can be used to identify the points, edges, objects, boundaries, and so on, described above. The classification layer can be used to identify cognitive states, mental states, emotional states, moods, and the like. The output of the final classification layer can be indicative of the emotional states of faces within the images, where the images can be processed using the deep neural network.

FIG. 20 shows data collection including multiple devices and locations 2000. A computer is initialized for convolutional processing. Data, including image data and voice data, can be obtained for analysis using convolutional image processing. The data can be obtained from multiple devices, settings, and locations. Images of an individual can be obtained in a number of different settings, including that of an occupant in a vehicle using an imaging device within a vehicle. A computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer.

The multiple mobile devices, vehicles, and locations 2000 can be used separately or in combination to collect video data on a user 2010. The video data can include facial data, image data, etc. Other data such as audio data, physiological data, and so on, can be collected on the user. While one person is shown, the video data, or other data, can be collected on multiple people. A user 2010 can be observed as she or he is performing a task, experiencing an event, viewing a media presentation, and so on. The user 2010 can be shown one or more media presentations, political presentations, social media, or another form of displayed media. The one or more media presentations can be shown to a plurality of people. The media presentations can be displayed on an electronic display coupled to a client device. The data collected on the user 2010 or on a plurality of users can be in the form of one or more videos, video frames, still images, etc. The plurality of videos can be of people who are experiencing different situations. Some example situations can include the user or plurality of users being exposed to TV programs, movies, video clips, social media, social sharing, and other such media. The situations could also include exposure to media such as advertisements, political messages, news programs, and so on. As noted before, video data can be collected on one or more users in substantially identical or different situations and viewing either a single media presentation or a plurality of presentations. The data collected on the user 2010 can be analyzed and viewed for a variety of purposes including expression analysis, mental state analysis, cognitive state analysis, and so on. The electronic display can be on a smartphone 2020 as shown, a tablet computer 2030, a personal digital assistant, a television, a mobile monitor, or any other type of electronic device. In one embodiment, expression data is collected on a mobile device such as a cell phone 2020, a tablet computer 2030, a laptop computer, or a watch. Thus, the multiple sources can include at least one mobile device, such as a phone 2020 or a tablet 2030, or a wearable device such as a watch or glasses (not shown). A mobile device can include a front-facing camera and/or a rear-facing camera that can be used to collect expression data. Sources of expression data can include a webcam, a phone camera, a tablet camera, a wearable camera, and a mobile camera. A wearable camera can comprise various camera devices, such as a watch camera. In addition to using client devices for data collection from the user 2010, data can be collected in a house 2040 using a web camera or the like; in a vehicle 2050 using a web camera, client device, etc.; by a social robot 2060, and so on.

As the user 2010 is monitored, the user 2010 might move due to the nature of the task, boredom, discomfort, distractions, or for another reason. As the user moves, the camera with a view of the user's face can be changed. Thus, as an example, if the user 2010 is looking in a first direction, the line of sight 2022 from the smartphone 2020 is able to observe the user's face, but if the user is looking in a second direction, the line of sight 2032 from the tablet 2030 is able to observe the user's face. Furthermore, in other embodiments, if the user is looking in a third direction, the line of sight 2042 from a camera in the house 2040 is able to observe the user's face, and if the user is looking in a fourth direction, the line of sight 2052 from the camera in the vehicle 2050 is able to observe the user's face. If the user is looking in a fifth direction, the line of sight 2062 from the social robot 2060 is able to observe the user's face. If the user is looking in a sixth direction, a line of sight from a wearable watch-type device, with a camera included on the device, is able to observe the user's face. In other embodiments, the wearable device is another device, such as an earpiece with a camera, a helmet or hat with a camera, a clip-on camera attached to clothing, or any other type of wearable device with a camera or other sensor for collecting expression data. The user 2010 can also use a wearable device including a camera for gathering contextual information and/or collecting expression data on other users. Because the user 2010 can move her or his head, the facial data can be collected intermittently when she or he is looking in a direction of a camera. In some cases, multiple people can be included in the view from one or more cameras, and some embodiments include filtering out faces of one or more other people to determine whether the user 2010 is looking toward a camera. All or some of the expression data can be continuously or sporadically available from the various devices and other devices.

The captured video data can include cognitive content, such as facial expressions, etc., and can be transferred over a network 2070. The network can include the Internet or other computer network. The smartphone 2020 can share video using a link 2024, the tablet 2030 using a link 2034, the house 2040 using a link 2044, the vehicle 2050 using a link 2054, and the social robot 2060 using a link 2064. The links 2024, 2034, 2044, 2054, and 2064 can be wired, wireless, and hybrid links. The captured video data, including facial expressions, can be analyzed on a cognitive state analysis engine 2080, on a computing device such as the video capture device, or on another separate device. The analysis could take place on one of the mobile devices discussed above, on a local server, on a remote server, and so on. In embodiments, some of the analysis takes place on the mobile device, while other analysis takes place on a server device. The analysis of the video data can include the use of a classifier. The video data can be captured using one of the mobile devices discussed above and sent to a server or another computing device for analysis. However, the captured video data including expressions can also be analyzed on the device which performed the capturing. The analysis can be performed on a mobile device where the videos were obtained with the mobile device and wherein the mobile device includes one or more of a laptop computer, a tablet, a PDA, a smartphone, a wearable device, and so on. In another embodiment, the analyzing comprises using a classifier on a server or another computing device different from the capture device. The analysis data from the cognitive state analysis engine can be processed by a cognitive state indicator 2090. The cognitive state indicator 2090 can indicate cognitive states, mental states, moods, emotions, etc. Further embodiments include inferring a cognitive state based on emotional content within a face detected within the facial image data, wherein the cognitive state includes of one or more of drowsiness, fatigue, distraction, impairment, sadness, stress, happiness, anger, frustration, confusion, disappointment, hesitation, cognitive overload, focusing, engagement, attention, boredom, exploration, confidence, trust, delight, disgust, skepticism, doubt, satisfaction, excitement, laughter, calmness, curiosity, humor, depression, envy, sympathy, embarrassment, poignancy, or mirth.

FIG. 21 is a system diagram for convolutional computing using a multilayered analysis engine. In the system 2100, a computer is initialized for convolutional neural network processing. A plurality of images for processing on the computer is obtained, using an imaging device. A multilayered analysis engine is trained on the computer, using the plurality of images. The multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer. A further image is evaluated, using the multilayered analysis engine. The evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion. The convolutional neural network analysis is output from the output layer.

The system 2100 can include network 2110, which can comprise the Internet, an intranet, or another computer network. Network 2110 can be used for communication among various machines. An image and audio collection machine 2120 has a memory 2126 which stores instructions and one or more processors 2124 coupled to the memory 2126, wherein the one or more processors 2124 can execute instructions. The image and audio collection machine 2120 can also have a network connection to carry cognitive state data 2160, and a display 2122 that can present cognitive state data, cognitive state profiles, mental state data, mental state profiles, emotional states, emotional state profiles, and so on. The image and audio collection machine 2120 can collect cognitive state data including image data, facial data, voice data, audio data, electrodermal data, biometric data, etc., from an occupant of a vehicle. In some embodiments, there are multiple image and audio collection machines 2120 that each collect cognitive state data including facial data. This type of collection machine can have a camera 2128 and/or a microphone 2130. In many embodiments, both a camera and a microphone will be present. Further embodiments include obtaining audio information and augmenting the analysis of the cognitive state data with the audio information. Once the cognitive state data 2160 has been collected, the image and audio collection machine 2120 can upload information to an evaluation or analysis machine 2140, based on the cognitive state data from an image of an individual. The image and audio collection machine 2120 can communicate with the analysis machine 2140 over the network 2110, the Internet, some other computer network, or by another method suitable for communication between two machines. In some embodiments, the analysis machine 2140 functionality is embodied in the image and audio collection machine 2120.

The analysis machine 2140 can have a network connection for cognitive state information 2162, cognitive state profile information, or cognitive state data, a memory 2146 which stores instructions, and one or more processors 2144 coupled to the memory 2146, wherein the one or more processors 2144 can execute instructions, and a display 2142. The analysis machine 2140 can receive cognitive state information, collected from an individual or multiple individuals, from the image and audio collection machine 2120, and can learn a cognitive state profile for the occupant. The analysis machine 2140 can also compare further cognitive state data with the cognitive state profile. The analysis machine 2140 can calculate facial expression metrics and cognitive state metrics. In some embodiments, the analysis machine 2140 also allows a user to view and evaluate the cognitive state data and cognitive state profiles for the individual or multiple individuals. The analysis machine 2140 can then provide cognitive state information and metrics 2164 to the outputting machine 2150 through network 2110. The cognitive state information, profile information, and metrics can be presented on a display 2152. In some embodiments, the image and audio collection machine 2120 can also function as the outputting machine 2150.

The outputting machine 2150 can have a memory 2156 which stores instructions, and one or more processors 2154 coupled to the memory 2156, wherein the one or more processors 2154 can execute instructions. The outputting machine can use a computer network 2110 (the Internet or another computer communication method) to request the cognitive state information 2162 from the analysis machine. The outputting machine 2150 can receive cognitive state information and metrics 2164, based on the cognitive state data 2160. The cognitive state information and metrics manipulated or tuned to be output a predetermined emotional recognition indication. In some embodiments, the outputting machine is set up to receive cognitive state data collected from an individual in a real-time or near real-time process. In other embodiments, the outputting machine is set up to receive the cognitive state data on an intermittent basis. In at least one embodiment, a single computer incorporates the image and audio collection machine, the analysis machine, and the outputting machine functionalities.

In embodiments, computer system for convolutional processing comprises: a memory which stores instructions; one or more processors coupled to the memory wherein the one or more processors, when executing the instructions which are stored, are configured to: initialize a computer for convolutional neural network processing; obtain, using an imaging device, a plurality of images for processing on the computer for convolutional processing; train, on the computer, a multilayered analysis engine, using the plurality of images, wherein the multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer; evaluate a further image, using the multilayered analysis engine, wherein the evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion; and output a convolutional neural network analysis from the at least one output layer.

In embodiments, a computer program product is embodied in a non-transitory computer readable medium for convolutional processing, the computer program product comprising code which causes one or more processors to perform operations of: initializing a computer for convolutional neural network processing; obtaining, using an imaging device, a plurality of images for processing on the computer for convolutional processing; training, on the computer, a multilayered analysis engine, using the plurality of images, wherein the multilayered analysis engine comprises multiple layers that include one or more convolutional layers, one or more hidden layers, and at least one output layer; evaluating a further image, using the multilayered analysis engine, wherein the evaluating includes identifying a facial portion and identifying a facial expression based on the facial portion; and outputting a convolutional neural network analysis from the at least one output layer.

Each of the above methods may be executed on one or more processors on one or more computer systems. Embodiments may include various forms of distributed computing, client/server computing, and cloud-based computing. Further, it will be understood that the depicted steps or boxes contained in this disclosure's flow charts are solely illustrative and explanatory. The steps may be modified, omitted, repeated, or re-ordered without departing from the scope of this disclosure. Further, each step may contain one or more sub-steps. While the foregoing drawings and description set forth functional aspects of the disclosed systems, no particular implementation or arrangement of software and/or hardware should be inferred from these descriptions unless explicitly stated or otherwise clear from the context. All such arrangements of software and/or hardware are intended to fall within the scope of this disclosure.

The block diagrams and flowchart illustrations depict methods, apparatus, systems, and computer program products. The elements and combinations of elements in the block diagrams and flow diagrams, show functions, steps, or groups of steps of the methods, apparatus, systems, computer program products and/or computer-implemented methods. Any and all such functions—generally referred to herein as a “circuit,” “module,” or “system”—may be implemented by computer program instructions, by special-purpose hardware-based computer systems, by combinations of special purpose hardware and computer instructions, by combinations of general purpose hardware and computer instructions, and so on.

A programmable apparatus which executes any of the above-mentioned computer program products or computer-implemented methods may include one or more microprocessors, microcontrollers, embedded microcontrollers, programmable digital signal processors, programmable devices, programmable gate arrays, programmable array logic, memory devices, application specific integrated circuits, or the like. Each may be suitably employed or configured to process computer program instructions, execute computer logic, store computer data, and so on.

It will be understood that a computer may include a computer program product from a computer-readable storage medium and that this medium may be internal or external, removable and replaceable, or fixed. In addition, a computer may include a Basic Input/Output System (BIOS), firmware, an operating system, a database, or the like that may include, interface with, or support the software and hardware described herein.

Embodiments of the present invention are neither limited to conventional computer applications nor the programmable apparatus that run them. To illustrate: the embodiments of the presently claimed invention could include an optical computer, quantum computer, analog computer, or the like. A computer program may be loaded onto a computer to produce a particular machine that may perform any and all of the depicted functions. This particular machine provides a means for carrying out any and all of the depicted functions.

Any combination of one or more computer readable media may be utilized including but not limited to: a non-transitory computer readable medium for storage; an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor computer readable storage medium or any suitable combination of the foregoing; a portable computer diskette; a hard disk; a random access memory (RAM); a read-only memory (ROM), an erasable programmable read-only memory (EPROM, Flash, MRAM, FeRAM, or phase change memory); an optical fiber; a portable compact disc; an optical storage device; a magnetic storage device; or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.

It will be appreciated that computer program instructions may include computer executable code. A variety of languages for expressing computer program instructions may include without limitation C, C++, Java, JavaScript™, ActionScript™, assembly language, Lisp, Perl, Tcl, Python, Ruby, hardware description languages, database programming languages, functional programming languages, imperative programming languages, and so on. In embodiments, computer program instructions may be stored, compiled, or interpreted to run on a computer, a programmable data processing apparatus, a heterogeneous combination of processors or processor architectures, and so on. Without limitation, embodiments of the present invention may take the form of web-based computer software, which includes client/server software, software-as-a-service, peer-to-peer software, or the like.

In embodiments, a computer may enable execution of computer program instructions including multiple programs or threads. The multiple programs or threads may be processed approximately simultaneously to enhance utilization of the processor and to facilitate substantially simultaneous functions. By way of implementation, any and all methods, program codes, program instructions, and the like described herein may be implemented in one or more threads which may in turn spawn other threads, which may themselves have priorities associated with them. In some embodiments, a computer may process these threads based on priority or other order.

Unless explicitly stated or otherwise clear from the context, the verbs “execute” and “process” may be used interchangeably to indicate execute, process, interpret, compile, assemble, link, load, or a combination of the foregoing. Therefore, embodiments that execute or process computer program instructions, computer-executable code, or the like may act upon the instructions or code in any and all of the ways described. Further, the method steps shown are intended to include any suitable method of causing one or more parties or entities to perform the steps. The parties performing a step, or portion of a step, need not be located within a particular geographic location or country boundary. For instance, if an entity located within the United States causes a method step, or portion thereof, to be performed outside of the United States then the method is considered to be performed in the United States by virtue of the causal entity.

While the invention has been disclosed in connection with preferred embodiments shown and described in detail, various modifications and improvements thereon will become apparent to those skilled in the art. Accordingly, the foregoing examples should not limit the spirit and scope of the present invention; rather it should be understood in the broadest sense allowable by law.

Convolutional computing using multilayered analysis engine (2024)
Top Articles
Latest Posts
Recommended Articles
Article information

Author: Jerrold Considine

Last Updated:

Views: 5701

Rating: 4.8 / 5 (78 voted)

Reviews: 93% of readers found this page helpful

Author information

Name: Jerrold Considine

Birthday: 1993-11-03

Address: Suite 447 3463 Marybelle Circles, New Marlin, AL 20765

Phone: +5816749283868

Job: Sales Executive

Hobby: Air sports, Sand art, Electronics, LARPing, Baseball, Book restoration, Puzzles

Introduction: My name is Jerrold Considine, I am a combative, cheerful, encouraging, happy, enthusiastic, funny, kind person who loves writing and wants to share my knowledge and understanding with you.