Title: An improved mean curvature-based bending model for cloth simulation
Authors: Xiaohui Tan; Ruiliang Guo
Addresses: Information Engineering College, Capital Normal University, No. 56, N. West 3rd Ring Road, Haidian District, Beijing, 100048, China ' School of Fashion, Beijing Institute of Fashion Technology, Beijing, 100028, China
Abstract: In cloth animation, the bending behaviour of a cloth is important for cloth simulation effects. The presentation of cloth bending properties plays a key role in cloth animation research because cloth is characterised by a strong resistance to stretch while weak resistance to bending. This paper proposed an improved approximate nonlinear bending model based on local geometric information. In the dynamic simulation, a cloth was divided into several regions according to mean curvature of surface. The bending force was updated according to the changes of the mean curvature in each region. Based on differential geometry and the dynamic simulation of the cloth, the calculation of bending force was simple and accurate with the proposed model. Experimental results show that wrinkles and folds generated in a natural way with the improved model and the efficiency of simulation is improved compared with the original algorithm.
Keywords: bending model; cloth simulation; mean curvature; surface segmentation; nonlinear; bending force; region partition.
DOI: 10.1504/IJITM.2021.114158
International Journal of Information Technology and Management, 2021 Vol.20 No.1/2, pp.95 - 109
Received: 23 Mar 2017
Accepted: 03 Oct 2017
Published online: 12 Apr 2021 *