AI Paper Atlas

Research map / 3D vision / 3D generation

SuperMat: Physically Consistent PBR Material Estimation at Interactive Rates

Yijia Hong, Yuan-Chen Guo, Ran Yi, Yulong Chen, Yan-Pei Cao, Lizhuang Ma

ICCV 2025 · 3D vision › 3D generation · approach: Score distillation for text-to-3D · 8 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add end-to-end training (trailing scheduler fix plus perceptual loss) (ablation)SuperMat · Image-space decomposition (192 objects), albedoPSNR: 24.26 -> 26.70Improved
add end-to-end training (trailing scheduler fix plus perceptual loss) (ablation)SuperMat · Image-space decomposition (192 objects), inference timetime (s): 3.09 -> 0.07Improved
add re-render loss (ablation)SuperMat · Image-space decomposition (192 objects), albedoPSNR: 26.70 -> 27.68Improved
add re-render loss (ablation)SuperMat · Image-space decomposition (192 objects), relightingPSNR: 26.41 -> 27.66Improved
add re-render loss (ablation)SuperMat · Image-space decomposition (192 objects), metallicPSNR: 24.54 -> 25.48Improved
use perceptual loss instead of SSI loss (ablation)SuperMat · Loss function comparison (Objaverse test, no re-render loss), albedoPSNR: 24.2359 -> 27.0082Improved
use perceptual loss instead of L1 loss (ablation)SuperMat · Loss function comparison (Objaverse test, no re-render loss), albedoPSNR: 26.8102 -> 27.0082Improved
train on synthetic plus real-captured data instead of real-captured only (ablation)SuperMat · Training data ablation (64 objects, 6144 inputs), albedoPSNR: 24.50 -> 27.58Improved
train on synthetic plus real-captured data instead of synthetic only (ablation)SuperMat · Training data ablation (64 objects, 6144 inputs), albedoPSNR: 27.41 -> 27.58Improved
train on synthetic plus real-captured data instead of synthetic only (ablation)SuperMat · Image-space decomposition (192 objects), metallicPSNR: 23.80 -> 23.77No clear effect
add UV refinement network to blended SuperMatMV UV maps (ablation)SuperMatMV pipeline · UV maps from SuperMatMV, albedoPSNR: 13.56 -> 23.92Improved
replace StableMaterialMV with SuperMat pipeline for 3D relightingUV backprojection pipeline · Novel-view relighting, BlenderVault (14 objects)PSNR: 23.57 -> 27.00Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy