MatDecompSDF: High-Fidelity 3D Shape and PBR Material Decomposition from Multi-View Images

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Chengyu, Bennett, Isabella, Scott, Henry, Zhang, Liang, Chen, Mei, Li, Hao, Zhao, Rui
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912792630525952
author Wang, Chengyu
Bennett, Isabella
Scott, Henry
Zhang, Liang
Chen, Mei
Li, Hao
Zhao, Rui
author_facet Wang, Chengyu
Bennett, Isabella
Scott, Henry
Zhang, Liang
Chen, Mei
Li, Hao
Zhao, Rui
contents We present MatDecompSDF, a novel framework for recovering high-fidelity 3D shapes and decomposing their physically-based material properties from multi-view images. The core challenge of inverse rendering lies in the ill-posed disentanglement of geometry, materials, and illumination from 2D observations. Our method addresses this by jointly optimizing three neural components: a neural Signed Distance Function (SDF) to represent complex geometry, a spatially-varying neural field for predicting PBR material parameters (albedo, roughness, metallic), and an MLP-based model for capturing unknown environmental lighting. The key to our approach is a physically-based differentiable rendering layer that connects these 3D properties to the input images, allowing for end-to-end optimization. We introduce a set of carefully designed physical priors and geometric regularizations, including a material smoothness loss and an Eikonal loss, to effectively constrain the problem and achieve robust decomposition. Extensive experiments on both synthetic and real-world datasets (e.g., DTU) demonstrate that MatDecompSDF surpasses state-of-the-art methods in geometric accuracy, material fidelity, and novel view synthesis. Crucially, our method produces editable and relightable assets that can be seamlessly integrated into standard graphics pipelines, validating its practical utility for digital content creation.
format Preprint
id arxiv_https___arxiv_org_abs_2507_04749
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MatDecompSDF: High-Fidelity 3D Shape and PBR Material Decomposition from Multi-View Images
Wang, Chengyu
Bennett, Isabella
Scott, Henry
Zhang, Liang
Chen, Mei
Li, Hao
Zhao, Rui
Computer Vision and Pattern Recognition
68U05
I.3.7; I.3.3; I.4.1
We present MatDecompSDF, a novel framework for recovering high-fidelity 3D shapes and decomposing their physically-based material properties from multi-view images. The core challenge of inverse rendering lies in the ill-posed disentanglement of geometry, materials, and illumination from 2D observations. Our method addresses this by jointly optimizing three neural components: a neural Signed Distance Function (SDF) to represent complex geometry, a spatially-varying neural field for predicting PBR material parameters (albedo, roughness, metallic), and an MLP-based model for capturing unknown environmental lighting. The key to our approach is a physically-based differentiable rendering layer that connects these 3D properties to the input images, allowing for end-to-end optimization. We introduce a set of carefully designed physical priors and geometric regularizations, including a material smoothness loss and an Eikonal loss, to effectively constrain the problem and achieve robust decomposition. Extensive experiments on both synthetic and real-world datasets (e.g., DTU) demonstrate that MatDecompSDF surpasses state-of-the-art methods in geometric accuracy, material fidelity, and novel view synthesis. Crucially, our method produces editable and relightable assets that can be seamlessly integrated into standard graphics pipelines, validating its practical utility for digital content creation.
title MatDecompSDF: High-Fidelity 3D Shape and PBR Material Decomposition from Multi-View Images
topic Computer Vision and Pattern Recognition
68U05
I.3.7; I.3.3; I.4.1
url https://arxiv.org/abs/2507.04749