Automatic human Facial Expressions Recognition (FER) is becoming of increased interest. FER finds its applications in many emerging areas such as affective computing and intelligent human computer interaction. Most of earlier works on FER rely on a combination of a powerful feature extractor and a supervised classifier. This way of proceeding however suffers from the intra-class variability of facial expression data. Moreover, it does not correctly scale with datasets containing emotions in the wild. Recent developments on image classification demonstrate the efficiency of deep learning in discovering multiple levels of representations, which provide greater robustness to intra-class variability and correctly scale with large datasets. The topic of the proposed session fits perfectly with the goals of IPTA 2019. It aims to present the most state-of-art advances in image processing for FER and bring expert insights on exploiting deep learning in this field.
The topics of interest include, but are not limited to:
(c) GdR 720 ISIS - CNRS - 2011-2019.