MeshFlow: Efficient Artistic Mesh Generation via MeshVAE and Flow-based Diffusion Transformer

2026-06-03

AI Summary

Existing autoregressive methods for 3D mesh generation suffer from low inference efficiency, limited scalability, and quantization errors introduced by discretizing vertex coordinates. To address these limitations, this work proposes a novel framework that integrates a contrastive learning–guided MeshVAE with a Rectified Flow Transformer, unifying the continuous vertex positions and discrete connectivity of meshes into a compact, continuous latent space. This enables parallel generation of all geometric elements in a single forward pass. Notably, the approach is the first to combine a variational autoencoder with a flow-based diffusion Transformer for mesh synthesis, achieving substantial improvements in both generation speed and geometric fidelity. The method attains state-of-the-art performance on standard benchmarks and generates meshes up to 18 times faster than the fastest existing autoregressive approach.

Abstract

We present MeshFlow, a new method for generating artist-like 3D meshes. Current mesh generators often adopt Auto-Regressive (AR) next-token prediction, a natural choice given the discrete nature of mesh topology. However, AR methods scale poorly because the inference cost is quadratic in mesh size. They also require discretizing the vertex coordinates, which introduces quantization errors. To address these challenges, we introduce a Variational Autoencoder (VAE) that, supervised with a contrastive loss, represents both continuous vertex positions and discrete connectivity in a continuous latent space. This latent space is significantly more compact than prior token-based mesh representations. We then build a 3D generator based on a Rectified Flow transformer, generating all mesh vertices and edges in parallel. Our model generates meshes 18x faster than the fastest AR generator while also achieving excellent accuracy across standard mesh-generation metrics. Homepage: https://mesh-flow.github.io/, Code: https://github.com/facebookresearch/meshflow

Problem

• 3D mesh generation
• Auto-Regressive
• quantization error
• inference efficiency
• discrete mesh representation

Innovation

• MeshVAE
• Flow-based Diffusion Transformer
• Parallel Mesh Generation
• Continuous Latent Representation
• Contrastive Learning