A Dense Deformation Field for Facial Expression Analysis in Dynamic Sequences of 3D Scans

In this paper, we present a fully automatic approach for identity-independent facial expression recognition from 3D video sequences. Towards this goal, we propose a novel approach to extract a dense scalar field that represents the deformations between faces conveying different expressions. We extract relevant features from this deformation field using LDA and then train a dynamic model on these features using HMM. Experiments conducted on BU-4DFE dataset following state-of-the-art settings show the effectiveness of the proposed approach.