Fabric dynamic simulation based on 3D motion estimation

The reality effect is an important factor for fabric dynamic simulation. This paper proposes an approach based on fabric 3D motion estimation to improve the reality effect of fabric dynamic simulation. Due to the very non-rigid motion of fabric and the ill-posed problem of 3D motion estimation, this paper establish a framework of Markov random field (MRF) with constraints. By incorporating the motion prior constraints into the MRF, the motion smoothness features between local fabric regions is reflected. By minimizing the energy function of the MRF, 3D motion estimation of the fabric is achieved. Then, the recovered 3D coordinates of fabric feature points are used to adjust simulation computing process based on the mass-spring model, and simulation results with good reality effect are obtained. Experimental results for fabric dynamic simulation demonstrate the feasibility of the proposed approach.