This paper presents our response to the first interna- tional challenge on Facial Emotion Recognition and Analysis. 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 detect. The first kernel matrix is computed using Local Gabor Binary Pattern histograms and a histogram intersection 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 recently proposed SimpleMKL algorithm. SVM outputs are then averaged to exploit temporal information in the sequence. To eval- uate 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 in- formation 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 participants and try to explain why our method had the best performance during the FERA challenge.