Scaling Mesh Generation via Compressive Tokenization

Haohan Weng, Zibo Zhao, Biwen Lei, Xianghui Yang, Jian Liu, Zeqiang Lai, Zhuo Chen, Yuhong Liu, Jie Jiang, Chunchao Guo, Tong Zhang, Shenghua Gao, C.L. Philip Chen; Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR), 2025, pp. 11093-11103

Abstract

We propose a compressive yet effective mesh tokenization, Blocked and Patchified Tokenization (BPT), facilitating the generation of meshes exceeding 8k faces. BPT compresses mesh sequences by employing block-wise indexing and patch aggregation, reducing their length by approximately 75% compared to the vanilla sequences. This compression milestone unlocks the potential to utilize mesh data with significantly more faces, thereby enhancing detail richness and improving generation robustness. Empowered with the BPT, we have built a foundation mesh generative model training on scaled mesh data to support flexible control for point clouds and images. Our model demonstrates the capability to generate meshes with intricate details and accurate topology, achieving SoTA performance on mesh generation and reaching the level for direct product usage.

Related Material

[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Weng_2025_CVPR, author = {Weng, Haohan and Zhao, Zibo and Lei, Biwen and Yang, Xianghui and Liu, Jian and Lai, Zeqiang and Chen, Zhuo and Liu, Yuhong and Jiang, Jie and Guo, Chunchao and Zhang, Tong and Gao, Shenghua and Chen, C.L. Philip}, title = {Scaling Mesh Generation via Compressive Tokenization}, booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR)}, month = {June}, year = {2025}, pages = {11093-11103} }