References & Citations
Computer Science > Computer Vision and Pattern Recognition
Title: Alpha Invariance: On Inverse Scaling Between Distance and Volume Density in Neural Radiance Fields
(Submitted on 2 Apr 2024 (v1), last revised 17 Apr 2024 (this version, v2))
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.
Submission history
From: Haochen Wang [view email][v1] Tue, 2 Apr 2024 17:58:57 GMT (8472kb,D)
[v2] Wed, 17 Apr 2024 01:41:59 GMT (8472kb,D)
Link back to: arXiv, form interface, contact.