MeshAnything V2:
Artist-Created Mesh Generation
with Adjacent Mesh Tokenization

Yiwen Chen1,2, Yikai Wang3*, Yihao Luo4, Zhengyi Wang2,3,
Zilong Chen2,3, Jun Zhu2,3, Chi Zhang5*, Guosheng Lin1*
1S-Lab, Nanyang Technological University, 2Shengshu,
3Tsinghua University, 4Imperial College London, 5Westlake University
*Corresponding Authors.
PDF Paper arXiv arXiv Code Hugging Face Gradio
Demo Video

Equipped with the newly proposed Adjacent Mesh Tokenization (AMT), MeshAnything V2 significantly surpasses MeshAnything in both performance and efficiency. MeshAnything V2 generates Artist-Created Meshes (AM) up to 1600 faces aligned with given shapes. Combined with various 3D asset production pipelines, it efficiently achieves high-quality, highly controllable AM generation.

Illustration of Adjacent Mesh Tokenization

Pipeline Image

Unlike previous methods that use three vertices to represent a face, AMT uses a single vertex whenever possible.
When this is impossible, AMT adds a special token & and restarts.

Our experiments demonstrate that AMT reduces the token sequence length by half on average. Its compact, and well-structured sequence representation enhances sequence learning, thereby significantly improving both the efficiency and performance of mesh generation.

Abstract

We introduce MeshAnything V2, an autoregressive transformer that generates Artist-Created Meshes (AM) aligned to given shapes. It can be integrated with various 3D asset production pipelines to achieve high-quality, highly controllable AM generation. MeshAnything V2 surpasses previous methods in both efficiency and performance using models of the same size.

These improvements are due to our newly proposed mesh tokenization method: Adjacent Mesh Tokenization (AMT). Different from previous methods that represent each face with three vertices, AMT uses a single vertex whenever possible. Compared to previous methods, AMT requires about half the token sequence length to represent the same mesh in average. Furthermore, the token sequences from AMT are more compact and well-structured, fundamentally benefiting mesh generation. Our extensive experiments show that AMT significantly improves the efficiency and performance of mesh generation.

More Results

Teaser Image

BibTeX

@misc{chen2024meshanythingv2artistcreatedmesh,
      title={MeshAnything V2: Artist-Created Mesh Generation With Adjacent Mesh Tokenization},
      author={Yiwen Chen and Yikai Wang and Yihao Luo and Zhengyi Wang and Zilong Chen and Jun Zhu and Chi Zhang and Guosheng Lin},
      year={2024},
      eprint={2408.02555},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2408.02555},
}