Skip to main content
Integrative Graduate
Education and
Research Traineeship
Home »
About IGERT »
Find an IGERT »
Showcase »
Highlights
IGERT Stories
Achievements
Library »
Conferences »
Highlight Photo
A Psychologically Inspired Model for Video-Based Facial Expression Recognition
Fig. 1: Non-verbal communication between a user and an embodied agent.
Credits:
AVEC
competition Organizers
Back to Highlight
More Highlight Photos
Click on image for larger version.
Fig. 2: Current algorithms would struggle with the above frames.
Figure 3: Clips from the same video. (Top) Only a single frame is needed to describe this clip. (Bottom) Many frames are needed to describe the expressions in this clip.
Figure 4: Results on the AVEC2011 development set.