# An evaluation of SVBRDF Prediction from Generative Image Models for Appearance Modeling of 3D Scenes

Source: [An evaluation of SVBRDF Prediction from Generative Image Models for Appearance Modeling of 3D Scenes](<https://www.sota2.com/research/paper/694238e845614693628503d5>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-18T05:00:04.345Z

Authors: Alban Gauthier, Valentin Deschaintre, Alexandre Lanvin, Fredo Durand, Adrien Bousseau, George Drettakis

Published: 2025-12-15T23:06:17.000Z

## Abstract

Digital content creation is experiencing a profound change with the advent of deep generative models. For texturing, conditional image generators now allow the synthesis of realistic RGB images of a 3D scene that align with the geometry of that scene. For appearance modeling, SVBRDF prediction networks recover material parameters from RGB images. Combining these technologies allows us to quickly generate SVBRDF maps for multiple views of a 3D scene, which can be merged to form a SVBRDF texture atlas of that scene. In this paper, we analyze the challenges and opportunities for SVBRDF prediction in the context of such a fast appearance modeling pipeline. On the one hand, single-view SVBRDF predictions might suffer from multiview incoherence and yield inconsistent texture atlases. On the other hand, generated RGB images, and the different modalities on which they are conditioned, can provide additional information for SVBRDF estimation compared to photographs. We compare neural architectures and conditions to identify designs that achieve high accuracy and coherence. We find that, surprisingly, a standard UNet is competitive with more complex designs. Project page: http://repo-sam.inria.fr/nerphys/svbrdf-evaluation

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [SVBRDF Multi-view Consistency on 5 synthetic scenes (test)](<https://www.sota2.com/research/sota/svbrdf-multi-view-consistency-on-5-synthetic-scenes-test.md>): SVBRDF Multi-view Consistency; 5 synthetic scenes (test); Basecolor Consistency: 0.056.

- [Basecolor Estimation on InteriorVerse (test)](<https://www.sota2.com/research/sota/basecolor-estimation-on-interiorverse-test.md>): BaseColor Estimation; Interiorverse (test); PSNR: 21.66.

- [Metallic Estimation on InteriorVerse 2633 images (test)](<https://www.sota2.com/research/sota/metallic-estimation-on-interiorverse-2633-images-test.md>): Metallic Estimation; InteriorVerse 2633 images (test); PSNR: 20.13.

- [Roughness Estimation on InteriorVerse 2633 images (test)](<https://www.sota2.com/research/sota/roughness-estimation-on-interiorverse-2633-images-test.md>): Roughness Estimation; InteriorVerse 2633 images (test); PSNR: 17.73.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.13950>)

[PDF](<https://arxiv.org/pdf/2512.13950.pdf>)
