TriFlow: Generating Artist-Like 3D Mesh Topology via Nearest-Vertex Vector Fields
About
We present TriFlow, a new generative approach for producing compact 3D meshes with artist-like triangle topology directly from input geometry conditions such as signed distance fields. Our key insight is to represent mesh topology as a nearest-vertex vector field (NVF) defined over the surface, where each point encodes its association to the nearest triangle vertex in the local barycentric frame. We train a latent flow-matching model to synthesize this field, enabling topology generation conditioned on the input geometry. To extract a coherent mesh, we cluster surface regions using the generated NVF and guide a constrained quadric error metric (QEM) mesh simplification with topology-aware optimization. This yields output meshes that closely match the input geometry while exhibiting structured, artist-like connectivity. Experiments demonstrate that TriFlow achieves stronger generalization and significantly improved topology quality compared to state-of-the-art learning-based approaches, alongside 90% lower Chamfer Distance and an 8x speedup.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Mesh Reconstruction | Objaverse (test) | Chamfer Distance0.12 | 10 | |
| Pairwise Preference Assessment | Perceptual Preference Dataset | Geometry Preference Rate97.9 | 8 | |
| Mesh Reconstruction Topology and Efficiency Analysis | TRELLIS | Normalized Error (‰)0.00e+0 | 5 | |
| 3D Mesh Reconstruction | TRELLIS generated shapes | CD0.2 | 5 | |
| Mesh Reconstruction Topology and Efficiency Analysis | Objaverse | NE (per mille)0.1 | 5 |