Log NeRF: Comparing Spaces for Learning Radiance Fields

TL;DR

This paper investigates the impact of different color spaces on NeRF training, demonstrating that using log RGB space improves rendering quality, robustness, and performance in low light conditions.

Contribution

It introduces the novel idea of training NeRFs in log RGB space, showing consistent improvements over traditional color spaces across various scenarios.

Findings

Log RGB space enhances rendering quality.

Training in log RGB improves robustness across scenes.

Log RGB performs well in low light conditions.

Abstract

Neural Radiance Fields (NeRF) have achieved remarkable results in novel view synthesis, typically using sRGB images for supervision. However, little attention has been paid to the color space in which the network is learning the radiance field representation. Inspired by the BiIlluminant Dichromatic Reflection (BIDR) model, which suggests that a logarithmic transformation simplifies the separation of illumination and reflectance, we hypothesize that log RGB space enables NeRF to learn a more compact and effective representation of scene appearance. To test this, we captured approximately 30 videos using a GoPro camera, ensuring linear data recovery through inverse encoding. We trained NeRF models under various color space interpretations linear, sRGB, GPLog, and log RGB by converting each network output to a common color space before rendering and loss computation, enforcing representation learning in different color spaces. Quantitative and qualitative evaluations demonstrate that using a log RGB color space consistently improves rendering quality, exhibits greater robustness across scenes, and performs particularly well in low light conditions while using the same bit-depth input images. Further analysis across different network sizes and NeRF variants confirms the generalization and stability of the log space advantage.