FlashMesh: Faster and Better Autoregressive Mesh Synthesis via Structured Speculation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912719237545984 |
|---|---|
| author | Shen, Tingrui Zhang, Yiheng Tang, Chen Ping, Chuan Zhao, Zixing Wan, Le Wang, Yuwang Wang, Ronggang He, Shengfeng |
| author_facet | Shen, Tingrui Zhang, Yiheng Tang, Chen Ping, Chuan Zhao, Zixing Wan, Le Wang, Yuwang Wang, Ronggang He, Shengfeng |
| contents | Autoregressive models can generate high-quality 3D meshes by sequentially producing vertices and faces, but their token-by-token decoding results in slow inference, limiting practical use in interactive and large-scale applications. We present FlashMesh, a fast and high-fidelity mesh generation framework that rethinks autoregressive decoding through a predict-correct-verify paradigm. The key insight is that mesh tokens exhibit strong structural and geometric correlations that enable confident multi-token speculation. FlashMesh leverages this by introducing a speculative decoding scheme tailored to the commonly used hourglass transformer architecture, enabling parallel prediction across face, point, and coordinate levels. Extensive experiments show that FlashMesh achieves up to a 2 x speedup over standard autoregressive models while also improving generation fidelity. Our results demonstrate that structural priors in mesh data can be systematically harnessed to accelerate and enhance autoregressive generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_15618 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FlashMesh: Faster and Better Autoregressive Mesh Synthesis via Structured Speculation Shen, Tingrui Zhang, Yiheng Tang, Chen Ping, Chuan Zhao, Zixing Wan, Le Wang, Yuwang Wang, Ronggang He, Shengfeng Computer Vision and Pattern Recognition Autoregressive models can generate high-quality 3D meshes by sequentially producing vertices and faces, but their token-by-token decoding results in slow inference, limiting practical use in interactive and large-scale applications. We present FlashMesh, a fast and high-fidelity mesh generation framework that rethinks autoregressive decoding through a predict-correct-verify paradigm. The key insight is that mesh tokens exhibit strong structural and geometric correlations that enable confident multi-token speculation. FlashMesh leverages this by introducing a speculative decoding scheme tailored to the commonly used hourglass transformer architecture, enabling parallel prediction across face, point, and coordinate levels. Extensive experiments show that FlashMesh achieves up to a 2 x speedup over standard autoregressive models while also improving generation fidelity. Our results demonstrate that structural priors in mesh data can be systematically harnessed to accelerate and enhance autoregressive generation. |
| title | FlashMesh: Faster and Better Autoregressive Mesh Synthesis via Structured Speculation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.15618 |