Facial Expression Recognition Based on Extended Optical Flow Constraint

Optical flow is an important method of motion image analysis and has a widely application in machine vision and image processing. The extended optical flow constraint based on the continuity equation of fluid motion is robust to the changes of brightness and deformation of motion objects. In this paper, on the basis of the extended optical flow constraint, a novel approach for estimating facial expression motion is presented. First-order and second-order div-curl splines as an additional constraint conditions are introduced for the calculation of non-rigid motion optical flow field, and the approach is used to calculated optical flow field of facial expressions sequences, The experiment results show the methods can compute effectively facial expression optical flow fields.