HUMAN BODY MODELLING USING QUADRATIC DEFORMATIONS
João K. Fayad, Alessio Del Bue, Lourdes Agapito, Pedro M. Q. Aguiar · 2009
The inference of the skeleton structure and joint locations of the human body is of focal interest in many applications such as clinical analysis and sport performance optimization. One of the major assumptions when dealing with this problem is that the body parts are rigid or quasi-rigid. However, soft tissue artifacts may appear with muscular contraction when observing the motion of the body. In this paper, we tackle this problem by proposing a more general shape model that accounts for quadratic deformations. Our approach takes Motion Capture (MOCAP) data as input and enables the extraction of more accurate estimates for the rigid component of the different body segments using a factorization framework. The parameters of the model are computed using a Levenberg-Marquardt non-linear optimization scheme. The proposed method can later be combined with other approaches that allow the construction of an articulated skeleton based on the rigid segments to create a model of the human body. The algorithm was extensively tested using synthetic data, and its performance assessed by comparisson with ground truth. We also present a qualitative analysis of the results obtained with MOCAP sequences.