Output details
15 - General Engineering
Brunel University London
Affective State Level Recognition in Naturalistic Facial and Vocal Expressions
This paper is an extended version of the conference paper “Naturalistic Affective Expression Classification by a Multi-Stage Approach Based on Hidden Markov Models” that won the sub-challenge of 1st International Audio/Visual Emotion Challenge (AVEC) in 2011. The proposed method achieved the best performance in AVEC2011 challenge contest for audio modality. In this paper, the method was improved, extended for both video and audio modality and evaluated in two more datasets. The research work has led to an international research exchange project (with Tsinghua University in China) funded by Royal society of Engineering for further development in 2012.