TensoRF: Tensorial Radiance Fields

doi:10.1007/978-3-031-19824-3_20

	TensoRF: Tensorial Radiance Fields
	Chen, Anpei1 ; Xu, Zexiang 2; Geiger, Andreas 3,4; Yu, Jingyi1 ; Su, Hao 5
	2022
会议录名称	COMPUTER VISION - ECCV 2022, PT XXXII
ISSN	0302-9743
卷号	13692
发表状态	已发表
DOI	10.1007/978-3-031-19824-3_20
摘要	We present TensoRF, a novel approach to model and reconstruct radiance fields. Unlike NeRF that purely uses MLPs, we model the radiance field of a scene as a 4D tensor, which represents a 3D voxel grid with per-voxel multi-channel features. Our central idea is to factorize the 4D scene tensor into multiple compact low-rank tensor components. We demonstrate that applying traditional CANDECOMP/PARAFAC (CP) decomposition - that factorizes tensors into rank-one components with compact vectors - in our framework leads to improvements over vanilla NeRF. To further boost performance, we introduce a novel vector-matrix (VM) decomposition that relaxes the low-rank constraints for two modes of a tensor and factorizes tensors into compact vector and matrix factors. Beyond superior rendering quality, our models with CP and VM decompositions lead to a significantly lower memory footprint in comparison to previous and concurrent works that directly optimize per-voxel features. Experimentally, we demonstrate that TensoRF with CP decomposition achieves fast reconstruction (< 30 min) with better rendering quality and even a smaller model size (< 4 MB) compared to NeRF. Moreover, TensoRF with VM decomposition further boosts rendering quality and outperforms previous state-of-the-art methods, while reducing the reconstruction time (< 10 min) and retaining a compact model size (< 75 MB).
会议名称	17th European Conference on Computer Vision (ECCV)
出版地	GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND
会议地点	null,Tel Aviv,ISRAEL
会议日期	OCT 23-27, 2022
URL	查看原文
收录类别	CPCI-S ; EI
语种	英语
WOS研究方向	Computer Science ; Imaging Science & Photographic Technology
WOS类目	Computer Science, Artificial Intelligence ; Imaging Science & Photographic Technology
WOS记录号	WOS:000903565400020
出版者	SPRINGER INTERNATIONAL PUBLISHING AG
EISSN	1611-3349
引用统计	被引频次：520[WOS] [WOS记录] [WOS相关记录]
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/272829
专题	信息科学与技术学院_博士生信息科学与技术学院_PI研究组_虞晶怡组
通讯作者	Chen, Anpei
作者单位	1.ShanghaiTech Univ, Shanghai, Peoples R China 2.Adobe Res, San Jose, CA USA 3.Univ Tubingen, Tubingen, Germany 4.MPI IS, Tubingen, Germany 5.Univ Calif San Diego, San Diego, CA USA
第一作者单位	上海科技大学
通讯作者单位	上海科技大学
第一作者的第一单位	上海科技大学
推荐引用方式 GB/T 7714	Chen, Anpei,Xu, Zexiang,Geiger, Andreas,et al. TensoRF: Tensorial Radiance Fields[C]. GEWERBESTRASSE 11, CHAM, CH-6330, SWITZERLAND:SPRINGER INTERNATIONAL PUBLISHING AG,2022.