AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

In-Place Scene Labelling and Understanding With Implicit Scene Representation

Shuaifeng Zhi, Tristan Laidlow, Stefan Leutenegger, Andrew J. Davison

ICCV 2021 · 3D vision › NeRF and radiance fields · approach: Sparse-input NeRF regularization · 437 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 multi-view consistent NeRF training to noisy pixel labels (90% flipped)Semantic-NeRF · Replica pixel-wise denoisingmIoU: 0.041 -> 0.877Improved
add Semantic-NeRF training to region-wise noisy labels (Even, 30%)Semantic-NeRF · Replica Room_2 region-wise denoisingmIoU: 0.741 -> 0.796Improved
add Semantic-NeRF training to region-wise noisy labels (Sort, 50%)Semantic-NeRF · Replica Room_2 region-wise denoisingmIoU: 0.793 -> 0.803No clear effect
use sparse geometrically accurate low-res labels instead of interpolated dense ones (S=8) (ablation)Semantic-NeRF label super-resolution · Replica label super-resolutionmIoU: 0.610 -> 0.887Improved
use sparse low-res labels instead of interpolated dense ones (S=16) (ablation)Semantic-NeRF label super-resolution · Replica label super-resolutionmIoU: 0.433 -> 0.800Improved
increase labelled pixels per class from single click to 10% (ablation)Semantic-NeRF label propagation · Replica label propagationmIoU: 0.602 -> 0.884Improved
replace Bayesian fusion with NeRF-training fusion of CNN predictionsmulti-view semantic fusion · Replica NYUv2-13 fusionmIoU: 0.668 -> 0.680Improved
add NeRF-training fusion to monocular CNN predictionsDeepLabV3+ · Replica NYUv2-13 fusionmIoU: 0.659 -> 0.680Improved
add average fusion to monocular CNN predictionsDeepLabV3+ · Replica NYUv2-13 fusionmIoU: 0.659 -> 0.586Hurt
add semantic prediction head to NeRF (ablation)NeRF · Replica reconstructionPSNR: 32.80 -> 32.27No clear effect
add semantic prediction head to NeRF (ablation)NeRF · Replica depthRMSE: 0.102 -> 0.096No clear effect
set semantic loss weight lambda to 1 instead of 0.04 (ablation)Semantic-NeRF · Replicasemantic performanceNo clear effect

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