DeferredGS: Decoupled and Relightable Gaussian Splatting With Deferred Shading

doi:10.1109/TPAMI.2025.3560933

	DeferredGS: Decoupled and Relightable Gaussian Splatting With Deferred Shading
	Wu, Tong 1,2; Sun, Jia-Mu 1,2; Lai, Yu-Kun 3; Ma, Yuewen 4; Kobbelt, Leif 5; Gao, Lin 1,2
	2025-08-01
发表期刊	IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
ISSN	0162-8828
卷号	47 期号:8 页码:6307-6319
摘要	Reconstructing and editing 3D objects and scenes both play crucial roles in computer graphics and computer vision. Neural radiance fields (NeRFs) can achieve realistic reconstruction and editing results but suffer from inefficiency in rendering. Gaussian splatting significantly accelerates rendering by rasterizing Gaussian ellipsoids. However, Gaussian splatting utilizes a single Spherical Harmonic (SH) function to model both texture and lighting, limiting independent editing capabilities of these components. Recently, attempts have been made to decouple texture and lighting with the Gaussian splatting representation but may fail to produce plausible geometry and decomposition results on reflective scenes. Additionally, the forward shading technique they employ introduces noticeable blending artifacts during relighting, as the geometry attributes of Gaussians are optimized under the original illumination and may not be suitable for novel lighting conditions. To address these issues, we introduce DeferredGS, a method for decoupling and relighting the Gaussian splatting representation using deferred shading. To achieve successful decoupling, we model the illumination with a learnable environment map and define additional attributes such as texture parameters and normal direction on Gaussians, where the normal is distilled from a jointly trained signed distance function. More importantly, we apply deferred shading, resulting in more realistic relighting effects compared to previous methods. Both qualitative and quantitative experiments demonstrate the superior performance of DeferredGSin novel view synthesis and relighting tasks.
关键词	Lighting Rendering (computer graphics) Geometry Three-dimensional displays Neural radiance field Training Solid modeling Surface reconstruction Harmonic analysis Image reconstruction Gaussian splatting inverse rendering editing
DOI	10.1109/TPAMI.2025.3560933
收录类别	SCI
语种	英语
资助项目	Beijing Municipal Science and Technology Commission[Z231100005923031] ; National Natural Science Foundation of China[62322210] ; Innovation Funding of ICT, CAS[E461020]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:001522958700047
出版者	IEEE COMPUTER SOC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/41759
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Gao, Lin
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing Key Lab Mobile Comp & Pervas Device, Beijing 100045, Peoples R China 2.Univ Chinese Acad Sci, Beijing 101408, Peoples R China 3.Cardiff Univ, Sch Comp Sci & Informat, Cardiff CF10 3AT, Wales 4.ByteDance Pico, Beijing 100098, Peoples R China 5.Rhein Westfal TH Aachen, D-52062 Aachen, Germany
推荐引用方式 GB/T 7714	Wu, Tong,Sun, Jia-Mu,Lai, Yu-Kun,et al. DeferredGS: Decoupled and Relightable Gaussian Splatting With Deferred Shading[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2025,47(8):6307-6319.
APA	Wu, Tong,Sun, Jia-Mu,Lai, Yu-Kun,Ma, Yuewen,Kobbelt, Leif,&Gao, Lin.(2025).DeferredGS: Decoupled and Relightable Gaussian Splatting With Deferred Shading.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,47(8),6307-6319.
MLA	Wu, Tong,et al."DeferredGS: Decoupled and Relightable Gaussian Splatting With Deferred Shading".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 47.8(2025):6307-6319.