We study the problem of few-shot physically-aware articulated mesh generation. By observing an articulated object dataset containing only a few examples, we wish to learn a model that can generate diverse meshes with high visual fidelity and physical validity. Previous mesh generative models either have difficulties in depicting a diverse data space from only a few examples or fail to ensure physical validity of their samples. Regarding the above challenges, we propose two key innovations, including 1) a hierarchical mesh deformation-based generative model based upon the divide-and-conquer philosophy to alleviate the few-shot challenge by borrowing transferrable deformation patterns from large scale rigid meshes and 2) a physics-aware deformation correction scheme to encourage physically plausible generations. We conduct extensive experiments on 6 articulated categories to demonstrate the superiority of our method in generating articulated meshes with better diversity, higher visual fidelity, and better physical validity over previous methods in the few-shot setting. Further, we validate solid contributions of our two innovations in the ablation study.
Our few-shot hierarchical mesh deformation scheme comprises three key designs: 1) learning and transferring common deformation patterns from large rigid datasets, 2) synchronizing convex deformations to consistent object deformations, and 3) improving the physical validity via a physics-aware deformation correction.
Please contact us at xymeow7@gmail.com if you have any question.
@inproceedings{liu2023fewshot,
title={Few-Shot Physically-Aware Articulated Mesh Generation via Hierarchical Deformation},
author={Liu, Xueyi and Wang, Bin and Wang, He and Yi, Li},
booktitle={International Conference on Computer Vision (ICCV)},
year={2023}
}