Keynote Speakers
Stephanie Schuckers
Bio. Dr. Stephanie Schuckers is the Bank of America Distinguished Professor in Computing & Informatics at University of North Caroline (UNC) – Charlotte. She also serves as the Director of the Center for Identification Technology Research (CITeR), a National Science Foundation Industry/University Cooperative Research Center, led by Clarkson University. She received her doctoral degree in Electrical Engineering from The University of Michigan. Professor Schuckers research focuses on processing and interpreting signals which arise from the human body. Her work is funded from various sources, including National Science Foundation, Department of Homeland Security, and private industry, among others. She has started her own business, testified for US Congress, and has over 50 journal publications as well as over 100 other academic publications. She was named IEEE Fellow in 2023, serves as a Board of Directors for the Biometrics Institute, and is President-Elect for the IEEE Biometrics Council. She has volunteered for numerous organizations including the IEEE Biometrics Council and FIDO Alliance.
Title. Security, Fairness, and Stability in Biometric Recognition: Challenges and Opportunities
Abstract. Biometric recognition has become an everyday part of life with applications from mobile devices to air travel to finance. There are two critical functions of biometric recognition (1) “matching” or determining if a sample matches a previous enrollment and (2) “presentation attack detection” or determining if the sample comes from a live individual present at the time of capture. Attacks include physical artefacts such as printouts, image/video display, or masks. More recently, with the leap in deepfake generation tools, digital injection attacks have become a more frequently exploited vulnerability. Injection attack detection (IAD) are a combination of cybersecurity protections (e.g. virtual camera detection) and biometric-related solutions, such as challenge response and deepfake detection. Studies have shown that errors related to PAD/IAD often overwhelm matching errors, suggesting that more research and evaluation are needed. This talk gives an overview of the field, discusses related research in fairness and explainability, and presents a longitudinal study of faces in children.
Jonathan Gratch
Bio. Jonathan Gratch is a Research Full Professor of Computer Science and Psychology at the University of Southern California (USC) and Director for Virtual Human Research at USC’s Institute for Creative Technologies. He completed his Ph.D. in Computer Science at the University of Illinois in Urbana-Champaign in 1995. Dr. Gratch’s research focuses on computational models of human cognitive and social processes, especially emotion, and explores these models’ potential to advance psychological theory and shape human-machine interaction. He is the founding Editor-in-Chief (retired) of IEEE’s Transactions on Affective Computing, Associate Editor for Affective Science, Emotion Review, and former President of the Association for the Advancement of Affective Computing (AAAC). He is a Fellow of AAAI, AAAC, and the Cognitive Science Society.
Title. A social-functional view on the recognition and analysis of emotional expressions
Abstract. Despite consensus among emotion researchers that the social meaning of emotional expressions is contextual, other-directed, co-constructed and culturally dependent, computational methods largely rest on the assumption that expressions denote some internal state (e.g., emotion or pain) which can be recovered by an expression’s morphology or timing alone. For example, many papers at the FG conference adopt a classic detection perspective, in which observers annotate the presumed internal states revealed by an expression, then algorithms are trained to predict this mapping without access to the original context in which the expression was produced. This assumption is also implicit in government regulations, such as the EU’s AI Act which bans emotion recognition across many practical. Though social psychology theirs point to a broader perspective on expressions, they fail to offer detailed of what constitutes “context” or “co-construction” to a level that can be exploited by computational methods. In this talk, I will highlight the use of automatic expression analysis in social domains, highlighting the interpersonal processes that shape their production and analysis. I hope this talk can encourage future work that formalizes the computational implications of this social perspective, including clarifying the communicative function of expressions, how are they shaped and co-constructed via context and culture, and how socially interactive agents might adapt and engage in meaning creation?
Kevin Bowyer
Bio. Kevin Bowyer is the Schubmehl-Prein Family Professor of Computer Science and Engineering at the University of Notre Dame. Professor Bowyer was elected as a Fellow of the American Academy for the Advancement of Science “for distinguished contributions to the field of computer vision and pattern recognition, biometrics, object recognition and data science”, a Fellow of the IEEE “for contributions to algorithms for recognizing objects in images”, and a Fellow of the IAPR “for contributions to computer vision, pattern recognition and biometrics”. He has received a Technical Achievement Award from the IEEE Computer Society “for pioneering contributions to the science and engineering of biometrics”, and IEEE Biometrics Council’s Meritorious Service Award and Leadership Award. Professor Bowyer has served as Editor-In-Chief of both the IEEE Transactions on Biometrics, Behavior, and Identity Science and the IEEE Transactions on Pattern Analysis and Machine Intelligence. Professor Bowyer has served as General Chair or Program Chair of conferences such as Computer Vision and Pattern Recognition, Winter Conference on Applications of Computer Vision, and Face and Gesture Recognition, and is one of the founding General Chairs of the International Joint Conference on Biometrics conference series.
Title. Face Recognition, Demographic Accuracy Variation, and Wrongful Arrest
Abstract. This talk first examines the issue of how face recognition accuracy is different across demographics. Then we consider the common assumption that demographic accuracy variation arises due to demographic imbalance in the training data. Finally, we evaluate the role of automated face recognition in high-profile instances of wrongful arrest. In each area, empirical evidence may not support some popular viewpoints.