LDM: Large tensorial SDF model for textured mesh generation

doi:10.1016/j.gmod.2025.101271

	LDM: Large tensorial SDF model for textured mesh generation
	Xie, Rengan 1; Huang, Kai 4; Luo, Xiaoliang 5; Chen, Yizheng 2; Wang, Lvchun 5; Wang, Qi 1; Ye, Qi 3; Chen, Wei 1; Zheng, Wenting 1; Huo, Yuchi 1,2
	2025-08-01
发表期刊	GRAPHICAL MODELS
ISSN	1524-0703
卷号	140 页码:13
摘要	Previous efforts have managed to generate production-ready 3D assets from text or images. However, these methods primarily employ NeRF or 3D Gaussian representations, which are not adept at producing smooth, high-quality geometries required by modern rendering pipelines. In this paper, we propose LDM, a Large tensorial SDF Model, which introduces a novel feed-forward framework capable of generating high-fidelity, illumination-decoupled textured mesh from a single image or text prompts. We firstly utilize a multi-view diffusion model to generate sparse multi-view inputs from single images or text prompts, and then a transformer-based model is trained to predict a tensorial SDF field from these sparse multi-view image inputs. Finally, we employ a gradient-based mesh optimization layer to refine this model, enabling it to produce an SDF field from which high-quality textured meshes can be extracted. Extensive experiments demonstrate that our method can generate diverse, high-quality 3D mesh assets with corresponding decomposed RGB textures within seconds. The project code is available at https://github.com/rgxie/LDM.
关键词	3D generation model Diffusion Intrinsic decomposition Relighting
DOI	10.1016/j.gmod.2025.101271
收录类别	SCI
语种	英语
资助项目	Zhejiang Province Jianbing Research and Development Project[2023C01042] ; National Key R&D Program of China[2023YFF0905102] ; Key Research and Development Program of Zhejiang Province[2025C01064] ; NSFC[U22B2034]
WOS研究方向	Computer Science
WOS类目	Computer Science, Software Engineering
WOS记录号	WOS:001520783400001
出版者	ACADEMIC PRESS INC ELSEVIER SCIENCE
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/42302
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Huo, Yuchi
作者单位	1.Zhejiang Univ, State Key Lab CAD&CG, Hangzhou, Peoples R China 2.Zhejiang Lab, Hangzhou, Peoples R China 3.Zhejiang Univ, State Key Lab Ind Control Technol, Hangzhou, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 5.Mobile Jiangxi Virtual Real Technol Co Ltd, Nanchang, Peoples R China
推荐引用方式 GB/T 7714	Xie, Rengan,Huang, Kai,Luo, Xiaoliang,et al. LDM: Large tensorial SDF model for textured mesh generation[J]. GRAPHICAL MODELS,2025,140:13.
APA	Xie, Rengan.,Huang, Kai.,Luo, Xiaoliang.,Chen, Yizheng.,Wang, Lvchun.,...&Huo, Yuchi.(2025).LDM: Large tensorial SDF model for textured mesh generation.GRAPHICAL MODELS,140,13.
MLA	Xie, Rengan,et al."LDM: Large tensorial SDF model for textured mesh generation".GRAPHICAL MODELS 140(2025):13.