Alpha Invariance: On Inverse Scaling Between Distance and Volume Density in Neural Radiance Fields
Abstract
Scale-ambiguity in 3D scene dimensions leads to magnitude-ambiguity of volumetric densities in neural radiance fields i.e. the densities double when scene size is halved and vice versa. We call this property alpha invariance. For NeRFs to better maintain alpha invariance we recommend 1) parameterizing both distance and volume densities in log space and 2) a discretization-agnostic initialization strategy to guarantee high ray transmittance. We revisit a few popular radiance field models and find that these systems use various heuristics to deal with issues arising from scene scaling. We test their behaviors and show our recipe to be more robust.
Cite
Text
Ahn et al. "Alpha Invariance: On Inverse Scaling Between Distance and Volume Density in Neural Radiance Fields." Conference on Computer Vision and Pattern Recognition, 2024. doi:10.1109/CVPR52733.2024.01928Markdown
[Ahn et al. "Alpha Invariance: On Inverse Scaling Between Distance and Volume Density in Neural Radiance Fields." Conference on Computer Vision and Pattern Recognition, 2024.](https://mlanthology.org/cvpr/2024/ahn2024cvpr-alpha/) doi:10.1109/CVPR52733.2024.01928BibTeX
@inproceedings{ahn2024cvpr-alpha,
title = {{Alpha Invariance: On Inverse Scaling Between Distance and Volume Density in Neural Radiance Fields}},
author = {Ahn, Joshua and Wang, Haochen and Yeh, Raymond A. and Shakhnarovich, Greg},
booktitle = {Conference on Computer Vision and Pattern Recognition},
year = {2024},
pages = {20396-20405},
doi = {10.1109/CVPR52733.2024.01928},
url = {https://mlanthology.org/cvpr/2024/ahn2024cvpr-alpha/}
}