Video-based online face recognition using identity surfaces

Abstract

Recognising faces across multiple views is more challenging than that from a fixed view because of the severe non-linearity caused by rotation in depth, self-occlusion, self-shading, and change of illumination. The problem can be related to the problem of modelling the spatiotemporal dynamics of moving faces from video input for unconstrained live face recognition. Both problems remain largely under-developed. To address the problems, a novel approach is presented in this paper. A multi-view dynamic face model is designed to extract the shape-and-pose-free texture patterns of faces. The model provides a precise correspondence to the task of recognition since the 3D shape information is used to warp the multi-view faces onto the model mean shape in frontal-view. The identity surface of each subject is constructed in a discriminant feature space from a sparse set of face texture patterns, or more practically, from one or more learning sequences containing the face of the subject. Instead of matching templates or estimating multi-modal density functions, face recognition can be performed by computing the pattern distances to the identity surfaces or trajectory distances between the object and model trajectories. Experimental results depict that this approach provides an accurate recognition rate while using trajectory distances achieves a more robust performance since the trajectories encode the spatio-temporal information and contain accumulated evidence about the moving faces in a video input.