而傳統的3D模型為輔助人臉辨識系統在對任一角度的人臉做角度檢索搜尋時耗時相當的久,所以本篇論文提出搭配粒子群最佳化的演算法來加速人臉角度檢索的步驟。 Face recognition has been one of the most popular topic of image processing. But there always exist a problem in conventional 2D face recognition system. It is the limitation of tolerance from face poses. For solving this problem, 3D model based face recognition has been proposed in recent years.
But it took a long time while matching the angles from an random head pose in conventional 3D model based face recognition system. So we accelerated the angle matching step by combining the Particle Swarm Optimization with 3D model based face recognition system.