Machine learning approaches have produced some of the highest reported performances for facial expression recognition. However, to date, nearly all automatic facial expression recognition research has focused on optimizing performance on a few databases that were collected under controlled lighting conditions on a relatively small number of subjects. This paper explores whether current machine learning methods can be used to develop an expression recognition system that operates reliably in more realistic conditions. We explore the necessary characteristics of the training data set, image registration, feature representation, and machine learning algorithms. A new database, GENKI, is presented which contains pictures, photographed by the subjects themselves, from thousands of different people in many different real-world imaging conditions. Results suggest that human-level expression recognition accuracy in real-life illumination conditions is achievable with machine learning technology. However, the data sets currently used in the automatic expression recognition literature to evaluate progress may be overly constrained and could potentially lead research into locally optimal algorithmic solutions.
}, keywords = {Algorithms, Artificial intelligence, Automated, automatic facial expression recognition research, Biological Pattern Recognition, Biometry, Computer simulation, Computer vision, Computer-Assisted, Face, Face and gesture recognition, face recognition, feature representation, human-level expression recognition accuracy, illumination conditions, Image databases, Image Enhancement, Image Interpretation, image registration image representation, learning (artificial intelligence), machine learning approaches, Machine Learning Models, n Humans, object detection, practical smile detection, Reproducibility of Results, Sensitivity and Specificity, Smiling, Subtraction Technique, training data set, visual databases}, issn = {0162-8828}, author = {Whitehill, J. and Littlewort, G. and Fasel, I. and Bartlett, M. and Movellan, J.} } @conference {62, title = {A discriminative approach to frame-by-frame head pose tracking}, booktitle = {8th IEEE International Conference on Automatic Face Gesture Recognition, 2008. FG {\textquoteright}08}, year = {2008}, month = {09/2008}, publisher = {IEEE}, organization = {IEEE}, address = {Amsterdam}, abstract = {We present a discriminative approach to frame-by-frame head pose tracking that is robust to a wide range of illuminations and facial appearances and that is inherently immune to accuracy drift. Most previous research on head pose tracking has been validated on test datasets spanning only a small (\< 20) subjects under controlled illumination conditions on continuous video sequences. In contrast, the system presented in this paper was both trained and tested on a much larger database, GENKI, spanning tens of thousands of different subjects, illuminations, and geographical locations from images on the Web. Our pose estimator achieves accuracy of 5.82deg, 5.65deg, and 2.96deg root-mean-square (RMS) error for yaw, pitch, and roll, respectively. A set of 4000 images from this dataset, labeled for pose, was collected and released for use by the research community.
}, keywords = {accuracy drift, continuous video sequence, controlled illumination condition, discriminative approach, face detection, face recognition, facial appearance, frame-by-frame head pose tracking, Humans, Image analysis, Image databases, Laboratories, Lighting, Magnetic heads, mean square error methods, pose estimation, Robustness, root-mean-square error tracking, System testing, Video sequences}, isbn = {978-1-4244-2153-4}, author = {Whitehill, J. and Movellan, Javier R.} }