English

A Neural Rendering Framework for Free-Viewpoint Relighting

Computer Vision and Pattern Recognition 2020-06-16 v2

Abstract

We present a novel Relightable Neural Renderer (RNR) for simultaneous view synthesis and relighting using multi-view image inputs. Existing neural rendering (NR) does not explicitly model the physical rendering process and hence has limited capabilities on relighting. RNR instead models image formation in terms of environment lighting, object intrinsic attributes, and light transport function (LTF), each corresponding to a learnable component. In particular, the incorporation of a physically based rendering process not only enables relighting but also improves the quality of view synthesis. Comprehensive experiments on synthetic and real data show that RNR provides a practical and effective solution for conducting free-viewpoint relighting.

Keywords

Cite

@article{arxiv.1911.11530,
  title  = {A Neural Rendering Framework for Free-Viewpoint Relighting},
  author = {Zhang Chen and Anpei Chen and Guli Zhang and Chengyuan Wang and Yu Ji and Kiriakos N. Kutulakos and Jingyi Yu},
  journal= {arXiv preprint arXiv:1911.11530},
  year   = {2020}
}

Comments

The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020

R2 v1 2026-06-23T12:27:39.298Z