AI Paper Atlas

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

Re-Distributing Biased Pseudo Labels for Semi-Supervised Semantic Segmentation: A Baseline Investigation

Ruifei He, Jihan Yang, Xiaojuan Qi

ICCV 2021 · Label-efficient and robust learning › Semi-supervised learning · approach: Consistency training for semi-supervised segmentation · 156 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 random sampling after distribution alignment (DARS vs DA+TS) (ablation)DA + TS · Cityscapes 1/8, crop 361mIoU: 67.31 -> 68.01Improved
replace single thresholding (ST) with DARS pseudo-labelingself-training · Cityscapes 1/8, crop 361mIoU: 65.70 -> 68.01Improved
replace class-balanced thresholding (CBST) with DARSself-training · Cityscapes 1/8, crop 361mIoU: 66.29 -> 68.01Improved
add temperature scaling to ST (ablation)ST · Cityscapes 1/8, crop 361mIoU: 65.70 -> 65.89Improved
add temperature scaling to CBST (ablation)CBST · Cityscapes 1/8, crop 361mIoU: 66.29 -> 66.61Improved
add iterative training (2 rounds) to DARS (ablation)DARS · Cityscapes 1/8, crop 361mIoU: 68.01 -> 69.64Improved
add iterative training to CBST (ablation)CBST · Cityscapes 1/8, crop 361mIoU: 66.29 -> 67.20Improved
add stronger augmentation (0.2, 0.5) at round 1 without enlarging labeling ratio (ablation)DARS round 1 · Cityscapes 1/8, crop 361mIoU: 68.01 -> 67.54Hurt
add second self-training round with alpha 20% (ablation)DARS · Cityscapes 1/8, crop 361mIoU: 68.01 -> 68.27Improved
enlarge labeling ratio to alpha 50% at round 2 (ablation)DARS · Cityscapes 1/8, crop 361mIoU: 68.27 -> 68.93Improved
add stronger random scaling range (beta 0.2, 0.5) at round 2 with alpha 50% (ablation)DARS · Cityscapes 1/8, crop 361mIoU: 68.93 -> 69.64Improved
use alpha 60% instead of 50% at round 2 (ablation)DARS · Cityscapes 1/8, crop 361mIoU: 68.93 -> 68.75Hurt

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