FACE CONTOUR TRACKING IN VIDEO USING ACTIVE CONTOUR MODEL (MP-P6)

Author(s) : Xiong Bing (EEE,NTU, Singapore) Yu Wei (EEE,NTU, Singapore) Charoensak Charayaphan (EEE,NTU, Singapore)

Abstract : Face contour represents important information used for human face analysis in image and video. Here, we propose an improvement to the conventional active contour model, called Snake, to track complex face contour. We propose using Gradient Angular Difference (GAD) to constrain the Snake contour to the specific face region. The Snake method obtains the optimal contour by iteratively minimizing its energy function. The iteration process is time consuming. Therefore, we proposed a motion compensation method to accelerate the face contour tracking process using Snake method in image sequences. The technique is shown to greatly reduce the computation time needed to minimize Snake energy. The simulation results show that the accurate Snake contour can be detected with minimal computation.

Menu