Dr. Jonathan Gratch to Deliver Keynote at The 19th IEEE International Conference on Automatic Face and Gesture Recognition

      Affective Computing Lab, ICT [L-R] Dr. Mohammad Soleymani, Dr. Gale Lucas, PhD student Bin 
      Han, Dr. Jon Gratch, Intern Sushrita Rakshit, Intern Emma Clift, and Phd student Ala Tak.

ICT’s Dr. Jonathan Gratch, Director, Affective Computing Lab, will deliver the Keynote speech “Facial Expressions in Context: How Affective Science is Shaping Expression Recognition” at The 19th IEEE International Conference on Automatic Face and Gesture Recognition (May 26 – 30, 2025).

This event is co-sponsored by IEEE Biometrics Council and IEEE Computer Society and is considered the premier international forum for research in image and video-based face, gesture, and body movement recognition. Topics in discussion at the conference will include advances in fundamental computer vision, pattern recognition, and computer graphics; machine learning techniques relevant to face, gesture, and body motion; interdisciplinary research on behavioral analysis; new algorithms and applications.

Dr. Gratch’s keynote presentation “Facial Expressions in Context: How Affective Science is Shaping Expression Recognition” focuses on the following concepts:

Affective computing is a field that applies computer science methods to recognize, interpret, simulate and stimulate human emotion. Though relying on data-driven approaches like machine learning, the field is surprisingly beholden to emotion theory, though it often fails to adapt to the most recent theoretical developments. For example, many methods that interpret facial expressions still assume expressions are universal signals of underlying basic emotions. In contrast, contemporary emotion theory emphasizes how context shapes the production and interpretation of facial displays. In this talk, I will discuss how these contemporary theories are beginning to shape computational methods. I will review several recent projects from my lab that illustrate how automatic methods can yield insight into the interpretation and social function of emotional displays in interdependent tasks such as social dilemmas and negotiations.  I will then highlight ways these methods and findings can enrich affective science.

“I’ll also be discussing some past and recent findings from my students,” Dr. Gratch added, “On how best to see facial expressions as communicative “tools” to shape interaction rather than some ‘secret window into our soul’.  As well as arguing that the new EU Artificial Intelligence Act somewhat embodies this “window into our soul” perspective, and how, from my own academic research perspective, I view this as somewhat out of step with the science behind affective computing, and AI.”

//