Abstract

Studies in psychology show that not all facial regions are of importance in recognizing facial expressions and different facial re- gions make different contributions in various facial expressions. Moti- vated by this, a novel framework, named Feature Disentangling Machine (FDM), is proposed to effectively select active features characterizing fa- cial expressions. More importantly, the FDM aims to disentangle these selected features into non-overlapped groups, in particular, common fea- tures that are shared across different expressions and expression-specific features that are discriminative only for a target expression. Specifically, the FDM integrates sparse support vector machine and multi-task learn- ing in a unified framework, where a novel loss function and a set of con- straints are formulated to precisely control the sparsity and naturally disentangle active features. Extensive experiments on two well-known facial expression databases have demonstrated that the FDM outper- forms the state-of-the-art methods for facial expression analysis. More importantly, the FDM achieves an impressive performance in a cross- database validation, which demonstrates the generalization capability of the selected features.