Duration: May 2013 to July 2013
University: Université de Montréal
Guide: Dr. Yoshua Bengio
Project Description
The Emotion Recognition in the Wild Challenge 2013 aimed at developing a system to classify the emotions expressed by the primary human subject in the short video clips of acted scenes lasting approximately one-two seconds, including the audio track which may contain human voices as well as background music.
I was a member of Université de Montréal's team for the Emotion Recognition in the wild challenge. I worked on adapting Wang et al's activity recognition using dense trajectories pipeline for emotion recognition. I was also involved in generating a video-emotion label dataset from descriptive video service files of movies.