Media.Player

Automatic Facial Expression Recognition System Using Emotion Avatar

Technology Benefits

This exciting technology can be applied for use in human-computer interactions, human emotional state identification, security surveillance systems, and safety monitoring systems.    The University of California is actively seeking licensing partnerships for this technology. If you are interested in more information about this technology or its licensing, please contact Christopher Del Vecchio at cvecchio@ucr.edu or (951) 827-4967, and he will be more than happy to answer all your questions.  UC Reference No. 2011-882

Detailed Technology Description

University of California (UC) researchers have developed a facial expression system able to automatically classify human facial expression in videos.  This novel invention utilizes a system that detects faces from videos, registers the face images, and creates a single image representation for the whole video.  The facial features are extracted to create an Emotion Avatar Image (EAI) as a single good representation for each video or image sequence for emotion recognition. Finally, a Support Vector Machine (SVM) is used for classification. By utilizing a single representation for a facial expression session, UC researchers are able to transform the whole expression recognition problem from image sequence back to static image that is amenable for real-time application.

Others

Tech ID/UC Case

22931/2011-882-1

Related Cases

2011-882-1

*Abstract

Current facial recognition techniques are limited to analyzing the spatial and temporal information for every single frame of video.  The inherent challenge for facial expression recognition and predicting human emotion is the dilemma between rigid motion of the head pose and the non-rigid motion of facial muscles.  Current technology has a credible capacity to estimate head pose, however, difficulty arises estimating non-rigid motion of facial muscles with issues such as non-rigid morphing and person specific appearance.  

      

*Principal Investigator

Name: Bir Bhanu

Department:

Name: Songfan Yang

Department:

Country/Region

USA