AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

RePL: Pseudo-label Refinement for Semi-supervised LiDAR Semantic Segmentation

Donghyeon Kwon, Taegyu Park, Suha Kwak

ECCV 2026 · Label-efficient and robust learning › Semi-supervised learning · approach: Consistency training for semi-supervised segmentation · 1 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 refiner supervised loss L_rsup (ablation)RePL refiner · nuScenes-lidarseg val (1% labeled)mIoU: 50.9 -> 57.2Improved
add negative-learning loss L_runl for refiner (ablation)RePL refiner · nuScenes-lidarseg val (1% labeled)mIoU: 57.2 -> 58.7Improved
add mixed-scene refiner loss L_rmix (ablation)RePL refiner · nuScenes-lidarseg val (1% labeled)mIoU: 58.7 -> 60.0Improved
add symmetric cross-entropy for unlabeled loss (ablation)RePL student · nuScenes-lidarseg val (1% labeled)mIoU: 58.0 -> 60.0Improved
add random masking when training refiner (ablation)RePL refiner · nuScenes-lidarseg val (1% labeled)mIoU: 57.7 -> 60.0Improved
replace random error mask with teacher-student agreement heuristic (ablation)RePL refinement · nuScenes-lidarseg val (1% labeled)mIoU: 58.7 -> 60.0Improved
replace heuristic error mask with oracle mask (ablation)RePL refinement · nuScenes-lidarseg val (1% labeled)mIoU: 60.0 -> 67.3Improved
reconstruct unreliable voxels instead of discarding them (ablation)RePL · nuScenes-lidarseg val (1% labeled)mIoU: 53.5 -> 60.0Improved
set confidence percentile kappa to 0.4 rather than 0.2/0.6 (ablation)RePL · nuScenes-lidarseg val (1% labeled)mIoU: 55.1 / 58.4 -> 60.0Mixed
apply RePL to SphereFormer backbone (ablation)SphereFormer · nuScenes-lidarseg val (1% labeled)mIoU: 50.7 -> 56.5Improved
apply refinement to pseudo-labelsRePL · nuScenes-lidarseg 1% per-class pseudo-label IoUIoU: 14 of 16 classes improved; driv. surf. -1.47, sidewalk -2.40Mixed
compare RePL with IT2 on average mIoURePL · nuScenes-lidarseg avgmIoU: 69.3 -> 71.3Improved

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