AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Relieving Long-Tailed Instance Segmentation via Pairwise Class Balance

Yin-Yin He, Peizhen Zhang, Xiu-Shen Wei, Xiangyu Zhang, Jian Sun

CVPR 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Classifier rebalancing and margin losses · 19 citations (OpenAlex)

Check 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 PCB to SeesawSeesaw on LVIS v0.5 · LVIS v0.5 val, R-50-FPNAP_r: 19.9 -> 23.4Improved
Add PCB to SeesawSeesaw on LVIS v1.0 · LVIS v1.0 val, R-50-FPNAP: 26.8 -> 27.2Improved
Add PCB regularization (ablation)RFS on LVIS v0.5 · LVIS v0.5 val, R-50-FPNAP_r: 16.0 -> 18.9Improved
Replace PCB regularization with online label smoothing (ablation)RFS on LVIS v0.5 · LVIS v0.5 val, R-50-FPNAP_r: 16.0 -> 15.4Hurt
Add iterative learning paradigm without PCB (ablation)RFS on LVIS v0.5 · LVIS v0.5 val, R-50-FPNAP: 25.6 -> 26.5Improved
Replace the iterative paradigm with deep supervision (ablation)PCB regularization with RFS · LVIS v0.5 val, R-50-FPNAP: 27.7 -> 26.9Hurt
Increase EMA momentum (ablation)PCB with RFS · LVIS v0.5 val, R-50-FPNAP_r: 19.3 -> 22.1Improved
Increase PCB regularization coefficient from 0.4 to 1.0 (ablation)PCB with RFS · LVIS v0.5 val, R-50-FPNAP: 27.7 -> 26.6Hurt
Use prediction-dependent iterative refinement instead of the CrowdDet refinement module (ablation)PCB regularization with RFS · LVIS v0.5 val, R-50-FPNAP: 26.4 -> 27.7Improved
Add PCB to CECE on ImageNet-LT · ImageNet-LT, ResNeXt-50Few accuracy: 7.44 -> 21.97Improved

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