Real-time Detection of End-effectors for Marker-free Motion Capture System

C.-J. Park and I.-H. Lee (Korea)


Human/Machine Interfaces, Motion Capture, Detection


This paper presents real-time end-effectors detection method, which can detect robustly human's head, hands, and feet for marker-free motion capture system. To detect robustly human's end-effectors for various poses, we use blob feature models such as shape model, color model, and spatial model, and utilize weighting factors to each model. According to the clothing conditions of human, the proposed method adjusts weights for each model automatically. Experiments are conducted on a lot of people wearing various clothes, and we confirmed that the proposed method could detect end-effectors of various poses of a lot of people wearing arbitrary clothes in real time.

Important Links:

Go Back