PGlight-NeRF: advancing scene representation through physically-grounded lighting modeling

Kai Li; Chengming Zou

doi:10.1117/12.3033947

16 October 2024 PGlight-NeRF: advancing scene representation through physically-grounded lighting modeling

Kai Li, Chengming Zou

Author Affiliations +

Proceedings Volume 13291, Ninth International Symposium on Advances in Electrical, Electronics, and Computer Engineering (ISAEECE 2024); 132911K (2024) https://doi.org/10.1117/12.3033947
Event: Ninth International Symposium on Advances in Electrical, Electronics, and Computer Engineering (ISAEECE 2024), 2024, Changchun, China

Abstract

Neural radiance fields (NeRF) is an important work in the fields of novel views synthesis, representing scenes as 5D functions that model the color emitted by objects directly based on their positions and observed directions. The rendering approach of NeRF faces challenges in accurately capturing the rapid and complex variations in specular reflections on glossy surfaces. In the computer graphics, physically based rendering (PBR) facilitates the real-time rendering of realistic lighting effects. Drawing inspiration from this, we deeply analyze the reflectance render equation and propose PGlightNeRF to enhance the lighting representation of NeRF. Our study indicates that our approach can achieve training and rendering in a short timeframe, synthesizing high-quality specular reflections views.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Kai Li and Chengming Zou "PGlight-NeRF: advancing scene representation through physically-grounded lighting modeling", Proc. SPIE 13291, Ninth International Symposium on Advances in Electrical, Electronics, and Computer Engineering (ISAEECE 2024), 132911K (16 October 2024); https://doi.org/10.1117/12.3033947

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE