SeamCrafter: Enhancing Mesh Seam Generation for Artist UV Unwrapping via Reinforcement Learning

• URL: http://arxiv.org/abs/2509.20725v2
• Date: Fri, 26 Sep 2025 03:01:38 GMT
• Title: SeamCrafter: Enhancing Mesh Seam Generation for Artist UV Unwrapping via Reinforcement Learning
• Abstract summary: We introduce SeamCrafter, an autoregressive GPT-style seam generator conditioned on point encoder inputs.<n>SeamCrafter produces seams with substantially lower distortion and fragmentation than prior approaches.
• Score: 17.67515163123214
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Mesh seams play a pivotal role in partitioning 3D surfaces for UV parametrization and texture mapping. Poorly placed seams often result in severe UV distortion or excessive fragmentation, thereby hindering texture synthesis and disrupting artist workflows. Existing methods frequently trade one failure mode for another-producing either high distortion or many scattered islands. To address this, we introduce SeamCrafter, an autoregressive GPT-style seam generator conditioned on point cloud inputs. SeamCrafter employs a dual-branch point-cloud encoder that disentangles and captures complementary topological and geometric cues during pretraining. To further enhance seam quality, we fine-tune the model using Direct Preference Optimization (DPO) on a preference dataset derived from a novel seam-evaluation framework. This framework assesses seams primarily by UV distortion and fragmentation, and provides pairwise preference labels to guide optimization. Extensive experiments demonstrate that SeamCrafter produces seams with substantially lower distortion and fragmentation than prior approaches, while preserving topological consistency and visual fidelity.

Related papers

• WAVE: Reversing the Guidance Hierarchy for Coarse-to-Fine Guided Depth Super-Resolution [50.15623093332659]
  Guided depth super-resolution (GDSR) typically extracts RGB guidance features through convolutional hierarchies, inheriting their fine-to-coarse bias.<n>We present WAVE, which introduces a multi-level discrete wavelet transform (ML-DWT) as an explicit and interpretable feature-control mechanism.<n>WAVE exploits sub-bands to treat high- and low-frequency content separately, filtering at its source the misleading RGB color and texture cues.
  arXiv  Detail & Related papers  (2026-08-26T02:16:41Z)
• ExMesh++: From Multi-View Images to Relightable UV-PBR Mesh Assets via Topology-Adaptive Reconstruction and Decomposition [36.887001139824456]
  We present ExMesh++, a framework for reconstructing relightable UV-PBR mesh assets from multi-view images.<n>Experiments demonstrate competitive geometry accuracy, strong relighting performance, and direct usability of the exported assets in standard DCC.
  arXiv  Detail & Related papers  (2026-08-25T06:16:03Z)
• Hybrid-Domain Posterior Sampling for Inverse Problems via Latent Flow Matching [22.663919850054928]
  Hybrid-Domain Posterior Sampling (HDPS) is a decoupled inference framework that disentangles physical measurement consistency from semantic prior modeling.<n> Experiments on diverse inverse problems demonstrate that HDPS establishes a new state-of-the-art, successfully recovering the high-frequency structural precision that latent-only solvers inherently discard.
  arXiv  Detail & Related papers  (2026-08-01T08:46:12Z)
• SeamGen: Artist-Aligned UV Seam Generation via Graph Flow Matching [47.43404509262908]
  SeamGen is a generative model for UV seam generation that aligns with artist preferences and production requirements.<n>By learning priors from professional seam layout data, SeamGen produces UV layouts that better align with artist-authored preferences.
  arXiv  Detail & Related papers  (2026-07-14T05:52:52Z)
• Monocular Avatar Reconstruction via Cascaded Diffusion Priors and UV-Space Differentiable Shading [22.58255448808833]
  We present a framework to reconstruct high-fidelity, relightable 3D avatars from a single in-the-wild image.<n>Our method, trained on fewer than 100 real 3D scans, generates comprehensive, 4K-resolution PBR assets with superior realism and generalization.
  arXiv  Detail & Related papers  (2026-06-26T14:41:13Z)
• DreamUV: Unwrap Artist-like UV by End-to-End Flow Matching [16.305062860577557]
  We present DreamUV, an end-to-end learning framework that formulates UV unwrapping as a generative Flow Matching problem.<n>We evaluate DreamUV on a large-scale dataset of professionally authored UV layouts.
  arXiv  Detail & Related papers  (2026-06-21T11:41:32Z)
• Preference Score Distillation: Leveraging 2D Rewards to Align Text-to-3D Generation with Human Preference [69.34278282513593]
  Preference Score Distillation (PSD) is an optimization-based framework for human-aligned text-to-3D synthesis without 3D training data.<n>Our key insight stems from the incompatibility of pixel-level gradients.<n>We introduce an adaptive strategy to co-optimize preference scores and negative text embeddings.
  arXiv  Detail & Related papers  (2026-03-02T08:23:36Z)
• LaFiTe: A Generative Latent Field for 3D Native Texturing [72.05710323154288]
  Existing native approaches are sparse by the absence of a powerful and versatile representation, which severely limits the fidelity and generality of their generated textures.<n>We introduce LaFiTe, which generates high-quality textures constrained by a sparse color representation and UV parameterization.
  arXiv  Detail & Related papers  (2025-12-04T13:33:49Z)
• PartUV: Part-Based UV Unwrapping of 3D Meshes [33.100992979838814]
  PartUV is a part-based UV unwrapping pipeline that generates fewer, part-aligned charts while maintaining low distortion.<n>PartUV is evaluated across four diverse datasets, including man-made, CAD, AI-generated, and Common Shapes.
  arXiv  Detail & Related papers  (2025-11-20T18:58:39Z)
• Unsupervised Representation Learning for 3D Mesh Parameterization with Semantic and Visibility Objectives [1.2110331500527989]
  We present an unsupervised differentiable framework that augments standard geometry-preserving UV learning with semantic- and visibility-aware objectives.<n>We show that the proposed method produces UV atlases that better support texture generation and reduce perceptible seam artifacts.
  arXiv  Detail & Related papers  (2025-09-29T17:28:58Z)
• ArtUV: Artist-style UV Unwrapping [15.072561481603037]
  We introduce ArtUV, a fully automated, end-to-end method for generating artist-style UV unwrapping.<n>We simulate the professional UV mapping process by dividing it into two stages: surface seam prediction and artist-style UV parameterization.<n>Our method ensures semantic consistency and preserves topological structure, making the UV map ready for 2D editing.
  arXiv  Detail & Related papers  (2025-09-25T03:21:21Z)
• Learning Deblurring Texture Prior from Unpaired Data with Diffusion Model [92.61216319417208]
  We propose a novel diffusion model (DM)-based framework, dubbed ours, for image deblurring.<n>ours performs DM to generate the prior knowledge that aids in recovering the textures of blurry images.<n>To fully exploit the generated texture priors, we present the Texture Transfer Transformer layer (TTformer)
  arXiv  Detail & Related papers  (2025-07-18T01:50:31Z)
• Auto-Regressive Surface Cutting [16.593710516134472]
  We introduce SeamGPT, an auto-recanned model that generates cutting seams by mimicking professional models.<n>Our approach achieves exceptional performance on UV unwrapping benchmarks containing both manifold and non-manifold meshes.<n>It enhances existing 3D segmentation tools by providing clean boundaries for part decomposition.
  arXiv  Detail & Related papers  (2025-06-22T12:53:07Z)
• UV-free Texture Generation with Denoising and Geodesic Heat Diffusions [50.55154348768031]
  Seams, wasted UV space, and varying resolution over the surface are the most prominent issues of the standard UV-based processing mechanism of meshes. We propose to represent textures as coloured point-cloud colours generated by a denoising diffusion model constrained to operate on the surface of 3D meshes.
  arXiv  Detail & Related papers  (2024-08-29T17:57:05Z)
• CVT-xRF: Contrastive In-Voxel Transformer for 3D Consistent Radiance Fields from Sparse Inputs [65.80187860906115]
  We propose a novel approach to improve NeRF's performance with sparse inputs. We first adopt a voxel-based ray sampling strategy to ensure that the sampled rays intersect with a certain voxel in 3D space. We then randomly sample additional points within the voxel and apply a Transformer to infer the properties of other points on each ray, which are then incorporated into the volume rendering.
  arXiv  Detail & Related papers  (2024-03-25T15:56:17Z)
• Texture Generation on 3D Meshes with Point-UV Diffusion [86.69672057856243]
  We present Point-UV diffusion, a coarse-to-fine pipeline that marries the denoising diffusion model with UV mapping to generate high-quality texture images in UV space. Our method can process meshes of any genus, generating diversified, geometry-compatible, and high-fidelity textures.
  arXiv  Detail & Related papers  (2023-08-21T06:20:54Z)
• IntrinsicNeRF: Learning Intrinsic Neural Radiance Fields for Editable Novel View Synthesis [90.03590032170169]
  We present intrinsic neural radiance fields, dubbed IntrinsicNeRF, which introduce intrinsic decomposition into the NeRF-based neural rendering method. Our experiments and editing samples on both object-specific/room-scale scenes and synthetic/real-word data demonstrate that we can obtain consistent intrinsic decomposition results.
  arXiv  Detail & Related papers  (2022-10-02T22:45:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.