Día: Miércoles 20 de diciembre
Lugar: H-1002
Hora: 11:00
Title: The Face of Emotion: A computational model of the production and visual perception of facial expression of emotion
Abstract: We now have computer vision algorithms that can successfully detect low-level image features (such as, edges), recover the 3D structure and motion of objects, and provide a semantic label for them (e.g., a face, or John’s face). But how about higher level, abstract concepts like emotions? This talk will introduce the first algorithms to successfully identify the emotion categories people regularly use to communicate emotion. I will first summarize our research uncovering the image features used by the human visual system to recognize emotion in faces. I will then explain how these results can be used to define computer vision systems that can work “in the wild” (i.e., outside controlled, in-lab conditions). In doing so, I will show the novel finding that people regularly use more than 35 distinct facial expressions (not the six propound by Darwin). We will see that the major problem in computer vision is that of landmark detection, not recognition, as is typical of most modern systems.
Bio: Aleix M. Martinez is a Professor in the Department of Electrical and Computer Engineering at The Ohio State University (OSU), where he is the founder and director of the Computational Biology and Cognitive Science Lab. He is also affiliated with the Department of Biomedical Engineering and the Center for Cognitive and Brain Sciences, where he is a member of the executive committee. Aleix has served as an associate editor of several major journals devoted to vision and affect (PAMI, TAC, CVIU) and area chair for many top conferences (CVPR, ICCV). He was also a Program Chair for CVPR 2014. He is a member of NIH’s Cognition and Perception study section. More about him:
http://www2.ece.ohio-state.edu/~aleix/