A Markerless Body Motion Capture System for Character Animation Based on Multi-view Cameras

Abstract

A novel application system is proposed in this paper to achieve the generation of 3D character animation driven by markerless human body motion capture. The whole pipeline of the system consists of four parts: capturing motion data by multiple cameras, detecting 2D human body joints and estimating 3D joints, calculating bone transformation matrices, and generating character animation. Its main objective is to generate 3D skeleton and animation for 3D characters from multi-view images captured by ordinary cameras. The computation complexity of 3D skeleton reconstruction based on 3D vision is reduced accordingly to achieve the frame-by-frame motion capture. The experimental results show that our system is effective and efficient for capturing human action and animating 3D cartoon characters simultaneously.