基于Mean Shift的运动人体骨架重构方法

Skeleton Extraction of Moving Human Using Mean Shift Algorithm

提出一种基于Mean Shift的方法来追踪并重构运动人体的骨架结构。首先,手动地将运动目标的部分肢体标识出来,为每一部分设定Mean Shift跟踪模板。由于单纯应用原始Mean Shift没有利用人体信息,提出一种结合肤色模型和边缘信息的Mean Shift算法,可将运动人体的骨架运动准确跟踪。最后利用上面的跟踪结合各运动区域的运动信息,将人体骨架完整地重构出来。实验结果表明,该方法能够实时有效地对运动人体的骨架进行重构。

A method for skeleton extraction of moving human using Mean Shift Algorithm is proposed.First,extract parts of the human skeleton as the Mean Shift tracking model manually.Then,use an improved Mean Shift method to track parts of the human body.This improved method will fully utilize the information of the human skeleton model.Finally,extract the human skeleton combining the motion information of the human.The experimental results show that the method proposed in this paper can be real-time and effective.