邻近不动点时序稳定的运动捕获数据恢复算法

A Temporally Stable Motion Capture Data Recovery Method Based on Proximity Fixed Point

针对具有低秩性、噪声稀疏性和时序稳定性的人体运动捕获数据恢复问题,提出一种基于邻近不动点时序稳定的运动捕获数据恢复算法(PFP-TS).给出PFP-TS算法的详细推导过程,并证明其收敛性.再通过数值实验将PFP-TS算法与PFP算法、TSMC算法进行对比,验证本文提出算法的可行性和有效性.

For the human motion capture data recovery problem with properties of low rank structure,noise sparsity and temporal stability,in this paper we proposed a proximity fixed point temporal stability(PFP-TS)algorithm.We derived the PPF-TS algorithm in details and analyzed its convergence.By comparing PFP-TS with PFP and TSMC in numerical experiment part,we verified the feasibility and effectiveness of our algorithm.