PolyFlow: Continuous Topology Embedding Flow Matching for Artist-style Mesh Generation

Hugging Face Daily Papers Papers
mesh-generation flow-matching topology-embedder transformer continuous-representation generative-model

Summary

PolyFlow introduces a continuous mesh representation using a topology embedder and applies flow-matching with Transformers for parallel mesh generation, achieving faster inference and precise resolution control compared to autoregressive methods.

Autoregressive Transformers dominate high-quality mesh generation by producing artist-worthy topologies, yet their inherent sequential decoding induces substantial computational overhead, falling orders of magnitude slower than parallel generative models. On the other hand, while continuous diffusion and flow-matching methods support efficient parallel synthesis across a variety of domains, they cannot be directly applied to meshes: mesh connectivity is inherently discrete and incompatible with standard continuous noise injection and denoising operations. To resolve this fundamental incompatibility, we introduce a compact topology embedder that projects discrete mesh vertex positions and normals into continuous per-vertex embeddings, where the original discrete adjacency information can be faithfully recovered via spacetime distance thresholding. After pretraining and freezing this embedder, any raw mesh can be fully converted into a continuous per-vertex state space unifying position, normal, and implicit topological attributes. Built upon this novel continuous mesh representation, we present PolyFlow, a Transformer-based flow-matching framework that achieves fully parallel vertex state denoising conditioned on extracted point-cloud features. During inference, our model completes generation rapidly via an ODE solver, and supports explicit, precise control over output mesh resolution by directly specifying the target vertex count. Extensive evaluations on the Toys4K benchmark demonstrate that PolyFlow surpasses state-of-the-art autoregressive baselines in both Chamfer Distance and Hausdorff Distance.
Original Article
View Cached Full Text

Cached at: 07/01/26, 03:40 AM

Paper page - PolyFlow: Continuous Topology Embedding Flow Matching for Artist-style Mesh Generation

Source: https://huggingface.co/papers/2606.30673 Authors:

,

,

,

,

,

,

,

,

,

,

Abstract

PolyFlow introduces a continuous mesh representation using a topology embedder and applies flow-matching with Transformers for parallel mesh generation, achieving faster inference and precise resolution control compared to autoregressive methods.

Autoregressive Transformersdominate high-qualitymesh generationby producing artist-worthy topologies, yet their inherent sequential decoding induces substantial computational overhead, falling orders of magnitude slower than parallel generative models. On the other hand, while continuous diffusion andflow-matchingmethods support efficient parallel synthesis across a variety of domains, they cannot be directly applied to meshes: mesh connectivity is inherently discrete and incompatible with standard continuous noise injection and denoising operations. To resolve this fundamental incompatibility, we introduce a compacttopology embedderthat projects discrete mesh vertex positions and normals into continuous per-vertex embeddings, where the original discrete adjacency information can be faithfully recovered via spacetime distance thresholding. After pretraining and freezing this embedder, any raw mesh can be fully converted into a continuous per-vertex state space unifying position, normal, and implicit topological attributes. Built upon this novelcontinuous mesh representation, we present PolyFlow, a Transformer-basedflow-matchingframework that achieves fully parallel vertex state denoising conditioned on extracted point-cloud features. During inference, our model completes generation rapidly via anODE solver, and supports explicit, precise control over output mesh resolution by directly specifying the target vertex count. Extensive evaluations on the Toys4K benchmark demonstrate that PolyFlow surpasses state-of-the-art autoregressive baselines in bothChamfer DistanceandHausdorff Distance.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2606\.30673

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.30673 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.30673 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.30673 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

MeshFlow: Mesh Generation with Equivariant Flow Matching

Hugging Face Daily Papers

MeshFlow introduces an equivariant optimal-transport flow matching model for direct triangle mesh generation, achieving state-of-the-art quality while providing approximately 18x inference speedup over autoregressive methods.

Meshy T2: Fast Native Mesh Generation with Flow Matching

Hugging Face Daily Papers

Meshy T2 introduces a fast native mesh generation framework using flow matching and a vertex-set mesh VAE, achieving state-of-the-art geometric fidelity with end-to-end image-to-mesh generation in a median of 6 seconds, over an order of magnitude faster than autoregressive baselines.

LATO.2: Factorized 3D Mesh Generation with Vertex and Topology Flow

Hugging Face Daily Papers

LATO.2 is a factorized flow matching framework for 3D mesh generation that decomposes the process into a vertex flow followed by a connectivity flow conditioned on the vertices, achieving state-of-the-art geometric fidelity and connectivity quality.

Flow Duality and Source Geometry for Categorical Generation

arXiv cs.LG

This paper identifies a duality between continuous and discrete flow matching, showing that projecting continuous convex-interpolant paths via argmax yields discrete flows, and explores how different source geometries affect transition timing and generation quality.