一种快速多人脸跟踪算法

Fast Multiple Faces Tracking Algorithm

提出一个基于肤色的快速多人脸跟踪算法。利用多个CAMShift跟踪器实现多人脸跟踪,提出最优排序法和目标消除法解决多人脸跟踪过程中目标发生粘连重叠的问题;引入多辅助信息和表决制解决了相邻两帧中人脸的对应问题。为进一步提高整个算法的跟踪速度和鲁棒性,引入卡尔曼滤波器对目标进行预测。实验结果表明,该算法可实时稳健地实现多人脸跟踪。

A novel skin-like color based multiple faces tracking algorithm is proposed in this paper. Multiple CAMShift trackers are built to implement the multiple face tracking. The taxis and object eliminating strategies are proposed to solve the occlusion problems. Multiple accessory information and voting mechanism are introduced to match the tracked objects and target windows. Furthermore, in order to improve the tracking speed and robustness, Kalman filter is imported to predict the position of the object window. Experimental results indicate that the proposed algorithm can track multiple faces robustly in real-time.