AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields

Jonathan T. Barron, Ben Mildenhall, Matthew Tancik, Peter Hedman, Ricardo Martin-Brualla, Pratul P. Srinivasan

ICCV 2021 · 3D vision › NeRF and radiance fields · approach: Dynamic and editable NeRF · 2,200 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add integrated positional encoding (cone-casting features) (ablation)Mip-NeRF · Multiscale Blenderaverage error: 0.0186 -> 0.0114Improved
add integrated positional encoding (ablation)Mip-NeRF · Blender (single-scale)average error: 0.0173 -> 0.0161Improved
add pixel-area loss scaling (ablation)Mip-NeRF · Multiscale Blenderaverage error: 0.0139 -> 0.0114Improved
add pixel-area loss scalingNeRF · Multiscale Blenderaverage error: 0.0288 -> 0.0305Mixed
use centered pixelsNeRF + Area Loss · Multiscale Blenderaverage error: 0.0305 -> 0.0191Improved
use centered pixelsNeRF · Blender (single-scale)average error: 0.0194 -> 0.0178Improved
add misc. stability changes (ablation)Mip-NeRF · Multiscale Blenderaverage error: 0.0114 -> 0.0114No clear effect
replace two MLPs with a single multiscale MLP (ablation)Mip-NeRF · Multiscale Blendertraining time (hours): 3.40 -> 2.84Improved
replace two MLPs with a single multiscale MLP (ablation)Mip-NeRF · Blender (single-scale)PSNR: 32.71 -> 33.09Improved
add supersamplingMip-NeRF · Multiscale Blendermean PSNR: 34.46 -> 35.03Improved
use mip-NeRF instead of supersampled NeRFNeRF · Multiscale Blenderrender time (sec./MP): 55.52 -> 2.48Improved
use mip-NeRF cone tracing instead of NeRFNeRF · Multiscale Blenderaverage error: 60%Improved

Later papers that compared with this method

38 later accepted papers report Mip-NeRF as a baseline in their results tables; 28 of them report a better number on most of the results they share with it.

30 more not listed.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy