1 - Zhejiang University, China

2 - University of Maryland at College Park, USA

3 - Tencent

Results: Given the initial template of the cloth mesh and the target obstacle mesh, our network can predicate a plausible target 3D cloth mesh for general scenes. We highlight (a) the final cloth mesh wrapped around a bunny; (b) the draping jacket on a non-SMPL human body; (c) the t-shirt deformation on a SMPL human body; (d) a human dressed in a robe represented by 100K triangles. All predicted meshes are different from the datasets used for training. Our approach runs at 30 − 45fps on an NVIDIA GeForce RTX 3090 GPU.

Abstract

We present a novel mesh-based learning approach (N-Cloth) for plausible 3D cloth deformation prediction. Our approach is general and can handle cloth or obstacles represented by triangle meshes with arbitrary topology. We use graph convolution to transform the cloth and object meshes into a latent space to reduce the non-linearity in the mesh space. Our network can predict the target 3D cloth mesh deformation based on the state of the initial cloth mesh template and the target obstacle mesh. Our approach can handle complex cloth meshes with up to 100K triangles and scenes with various objects corresponding to SMPL humans, Non-SMPL humans, or rigid bodies. In practice, our approach demonstrates good temporal coherence between successive input frames and can be used to generate plausible cloth simulation at 30−45 fps on an NVIDIA GeForce RTX 3090 GPU. We highlight its benefits over prior learning-based methods and physically-based cloth simulators.
 

Contents