This paper presents our response to the first internatio- nal challenge on Facial Emotion Recognition and Analy- sis. We propose to combine different types of features to automatically detect Action Units in facial images. We use one multi-kernel SVM for each Action Unit we want to de- tect. The first kernel matrix is computed using Local Ga- bor Binary Pattern histograms and a histogram intersec- tion kernel. The second kernel matrix is computed from AAM coefficients and an RBF kernel. During the training step, we combine these two types of features using the recent SimpleMKL algorithm. SVM outputs are then fil- tered to exploit temporal information in the sequence. To evaluate our system, we perform deep experimentations on several key issues : influence of features and kernel function in histogram-based SVM approaches, influence of spatially-independent information versus geometric local appearance information and benefits of combining both, sensitivity to training data and interest of temporal context adaptation. We also compare our results to those of the other challengers and try to explain why our method had the best performance during the FERA challenge.
|Original language||English (US)|
|Title of host publication||RFIA 2012 (Reconnaissance des Formes et Intelligence Artificielle)|
|State||Published - 2012|