English

VDN-NeRF: Resolving Shape-Radiance Ambiguity via View-Dependence Normalization

Computer Vision and Pattern Recognition 2023-04-03 v1

Abstract

We propose VDN-NeRF, a method to train neural radiance fields (NeRFs) for better geometry under non-Lambertian surface and dynamic lighting conditions that cause significant variation in the radiance of a point when viewed from different angles. Instead of explicitly modeling the underlying factors that result in the view-dependent phenomenon, which could be complex yet not inclusive, we develop a simple and effective technique that normalizes the view-dependence by distilling invariant information already encoded in the learned NeRFs. We then jointly train NeRFs for view synthesis with view-dependence normalization to attain quality geometry. Our experiments show that even though shape-radiance ambiguity is inevitable, the proposed normalization can minimize its effect on geometry, which essentially aligns the optimal capacity needed for explaining view-dependent variations. Our method applies to various baselines and significantly improves geometry without changing the volume rendering pipeline, even if the data is captured under a moving light source. Code is available at: https://github.com/BoifZ/VDN-NeRF.

Keywords

Cite

@article{arxiv.2303.17968,
  title  = {VDN-NeRF: Resolving Shape-Radiance Ambiguity via View-Dependence Normalization},
  author = {Bingfan Zhu and Yanchao Yang and Xulong Wang and Youyi Zheng and Leonidas Guibas},
  journal= {arXiv preprint arXiv:2303.17968},
  year   = {2023}
}
R2 v1 2026-06-28T09:42:54.101Z