Neural 3D reconstruction from sparse views using geometric priors

Computational Visual Media

Research Article

Ning Guo¹  Jun-Xiong Cai²  Tai-Jiang Mu²  Hao-Xiang Chen² 

[1] Academy of Military Sciences, 100091, Beijing, China;BNRist, Department of Computer Science and Technology, Tsinghua University, 100084, Beijing, China;

关键词: sparse views;    3D reconstruction;    volume rendering;    geometric priors;    neural implicit 3D representation;   

DOI  :  10.1007/s41095-023-0337-5

 received in 2022-11-25, accepted in 2023-01-31,  发布年份 2023

来源: Springer

【 摘 要 】

Sparse view 3D reconstruction has attracted increasing attention with the development of neural implicit 3D representation. Existing methods usually only make use of 2D views, requiring a dense set of input views for accurate 3D reconstruction. In this paper, we show that accurate 3D reconstruction can be achieved by incorporating geometric priors into neural implicit 3D reconstruction. Our method adopts the signed distance function as the 3D representation, and learns a generalizable 3D surface reconstruction model from sparse views. Specifically, we build a more effective and sparse feature volume from the input views by using corresponding depth maps, which can be provided by depth sensors or directly predicted from the input views. We recover better geometric details by imposing both depth and surface normal constraints in addition to the color loss when training the neural implicit 3D representation. Experiments demonstrate that our method both outperforms state-of-the-art approaches, and achieves good generalizability.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

Fig. 1