FACE: A Face-based Autoregressive Representation for High-Fidelity and Efficient Mesh Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914364334800896 |
|---|---|
| author | Wang, Hanxiao Guo, Yuan-Chen Liu, Ying-Tian Zou, Zi-Xin Zhang, Biao Quan, Weize Liang, Ding Cao, Yan-Pei Yan, Dong-Ming |
| author_facet | Wang, Hanxiao Guo, Yuan-Chen Liu, Ying-Tian Zou, Zi-Xin Zhang, Biao Quan, Weize Liang, Ding Cao, Yan-Pei Yan, Dong-Ming |
| contents | Autoregressive models for 3D mesh generation suffer from a fundamental limitation: they flatten meshes into long vertex-coordinate sequences. This results in prohibitive computational costs, hindering the efficient synthesis of high-fidelity geometry. We argue this bottleneck stems from operating at the wrong semantic level. We introduce FACE, a novel Autoregressive Autoencoder (ARAE) framework that reconceptualizes the task by generating meshes at the face level. Our one-face-one-token strategy treats each triangle face, the fundamental building block of a mesh, as a single, unified token. This simple yet powerful design reduces the sequence length by a factor of nine, leading to an unprecedented compression ratio of 0.11, halving the previous state-of-the-art. This dramatic efficiency gain does not compromise quality; by pairing our face-level decoder with a powerful VecSet encoder, FACE achieves state-of-the-art reconstruction quality on standard benchmarks. The versatility of the learned latent space is further demonstrated by training a latent diffusion model that achieves high-fidelity, single-image-to-mesh generation. FACE provides a simple, scalable, and powerful paradigm that lowers the barrier to high-quality structured 3D content creation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_01515 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FACE: A Face-based Autoregressive Representation for High-Fidelity and Efficient Mesh Generation Wang, Hanxiao Guo, Yuan-Chen Liu, Ying-Tian Zou, Zi-Xin Zhang, Biao Quan, Weize Liang, Ding Cao, Yan-Pei Yan, Dong-Ming Computer Vision and Pattern Recognition Autoregressive models for 3D mesh generation suffer from a fundamental limitation: they flatten meshes into long vertex-coordinate sequences. This results in prohibitive computational costs, hindering the efficient synthesis of high-fidelity geometry. We argue this bottleneck stems from operating at the wrong semantic level. We introduce FACE, a novel Autoregressive Autoencoder (ARAE) framework that reconceptualizes the task by generating meshes at the face level. Our one-face-one-token strategy treats each triangle face, the fundamental building block of a mesh, as a single, unified token. This simple yet powerful design reduces the sequence length by a factor of nine, leading to an unprecedented compression ratio of 0.11, halving the previous state-of-the-art. This dramatic efficiency gain does not compromise quality; by pairing our face-level decoder with a powerful VecSet encoder, FACE achieves state-of-the-art reconstruction quality on standard benchmarks. The versatility of the learned latent space is further demonstrated by training a latent diffusion model that achieves high-fidelity, single-image-to-mesh generation. FACE provides a simple, scalable, and powerful paradigm that lowers the barrier to high-quality structured 3D content creation. |
| title | FACE: A Face-based Autoregressive Representation for High-Fidelity and Efficient Mesh Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.01515 |