Skip to yearly menu bar Skip to main content


Poster

Alpha Invariance: On Inverse Scaling Between Distance and Volume Density in Neural Radiance Fields

Joshua Ahn · Haochen Wang · Raymond A. Yeh · Greg Shakhnarovich

Arch 4A-E Poster #83
[ ] [ Project Page ] [ Paper PDF ]
[ Poster
Fri 21 Jun 10:30 a.m. PDT — noon PDT

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.

Chat is not available.