Detail?Aware Deep Clothing Animations Infused with Multi?Source Attributes

This paper presents a novel learning-based clothing deformation method to generate rich and reasonable detailed deformations for garments worn by bodies of various shapes in animations. In contrast existing methods, which require numerous trained models different garment topologies or poses are unable easily realize details, we use unified framework produce high fidelity efficiently easily. Specifically, first found that the fit between body has an important impact on degree folds. We then designed attribute parser detail-aware encodings infused them into graph neural network, therefore enhancing discrimination details under diverse attributes. Furthermore, achieve better convergence avoid overly smooth deformations, proposed reconstruct output mitigate complexity learning task. Experimental results show our achieves performance over methods terms generalization ability quality details.