Non-rigid point cloud registration module based on local human body parts

Human point cloud registration is a key step in computer graphics and computer vision applications. However, human point cloud registration is challenging due to the problems of joint hinges and adhesion of human parts. In order to solve the above problems, this paper proposes a non-rigid point cloud registration method based on local human body parts. Firstly, the SMPL model is used to extract the joint points of the human body, and then the human body is segmented into different parts. Finally, the registration is performed according to the human body parts. The proposed method is sensitive to local features and can greatly reduce the difficulty of non-rigid registration.