Paper page - SUPRA-TEX: 3D Generation with Super-Aligned Texture

Source: https://huggingface.co/papers/2610.22352

Abstract

SUPRA-Tex decouples visible-material estimation, surface alignment, and 3D completion to generate relightable 3D materials that preserve reference image details and complete unobserved surfaces.

Single image 3D generationaims to dress a given 3D mesh with relightable materials while faithfully preserving the visual details of a reference image. Existing methods either introduce artifacts when fusing inconsistent multi-view predictions or struggle to preserve fine details through implicit image-feature conditioning. To resolve these challenges, we introduce SUPRA-Tex, built on the principle of decoupling visible-material estimation, surface alignment, and observation-guided 3D completion. This division of labor allows the 3D generator, trained on limited 3D data, to focus its capacity on preserving visible evidence and completing unobserved surfaces. Specifically, we first train anMLLM-based image editingmodel to predict lighting-separatedPBR mapsin the reference view. We then project these maps onto the visible surface and encode them with the pretrainedmaterial VAE encoderinto nativeobservation latentsthat share the generation target’s latent space. Aninterleaved flow transformerholds these latents as fixed, read-only anchors while alternating unobserved-region and full-surface updates to generate complete material latents. This design achieves super-alignment with the reference, delivering What You See Is What You Get fidelity on visible regions and coherent completion on unobserved surfaces. Trained solely on public datasets, SUPRA-Tex achieves SOTA performance in bothgeometry-conditioned texturingand end-to-end asset generation, matching or even exceeding leading commercial systems on in-the-wild images. Project page: https://supra-3d.github.io/supratex/

View arXiv pageView PDFProject pageGitHub28Add to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.22352 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.22352 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.22352 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.