Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

EdgeRunner: Auto-regressive Auto-encoder for Artistic Mesh Generation

About

Current auto-regressive mesh generation methods suffer from issues such as incompleteness, insufficient detail, and poor generalization. In this paper, we propose an Auto-regressive Auto-encoder (ArAE) model capable of generating high-quality 3D meshes with up to 4,000 faces at a spatial resolution of $512^3$. We introduce a novel mesh tokenization algorithm that efficiently compresses triangular meshes into 1D token sequences, significantly enhancing training efficiency. Furthermore, our model compresses variable-length triangular meshes into a fixed-length latent space, enabling training latent diffusion models for better generalization. Extensive experiments demonstrate the superior quality, diversity, and generalization capabilities of our model in both point cloud and image-conditioned mesh generation tasks.

Jiaxiang Tang, Zhaoshuo Li, Zekun Hao, Xian Liu, Gang Zeng, Ming-Yu Liu, Qinsheng Zhang• 2024

Related benchmarks

TaskDatasetResultRank
3D Mesh GenerationObjaverse
Chamfer Distance0.053
18
Mesh Tokenization3D Mesh Representation
Compression Ratio0.47
12
Mesh TokenizationToys4K (test)
Compression Ratio0.47
8
Mesh TokenizationMesh Sequences
Compression Ratio0.47
7
Point-Cloud-Conditioned Mesh GenerationToys4k
Chamfer Distance (CD)0.147
6
Mesh ReconstructionObjaverse-like proprietary dataset 1.0 (test)
Chamfer Distance (CD)1.21
3
Showing 6 of 6 rows

Other info

Follow for update