AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Overcoming Classifier Imbalance for Long-Tail Object Detection With Balanced Group Softmax

Yu Li, Tao Wang, Bingyi Kang, Sheng Tang, Chunfeng Wang, Jintao Li, Jiashi Feng

CVPR 2020 · Detection and segmentation › Open-set and few-shot detection · approach: Open-set and unknown object detection · 283 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
Replace the standard classifier with balanced group softmaxFaster R-CNN R50-FPN · LVIS valmAP: 20.98 -> 25.96Improved
Replace the standard classifier with balanced group softmaxFaster R-CNN R50-FPN · LVIS valAP₁: 0.00 -> 11.33Improved
Use BAGSFaster R-CNN R50-FPN · LVIS valmAP: 20.98 -> 25.96Improved
Use BAGSCascade R-CNN X101-FPN · LVIS valmAP: 27.16 -> 32.77Improved
Use BAGSMask R-CNN R50-FPN · LVIS valmask mAP: 20.68 -> 26.25Improved
Use BAGSFaster R-CNN R50-FPN · COCO-LT valmAP: 20.3 -> 22.5Improved
Add a separate background group to group softmax (ablation)BAGS with four normal-category groups · LVIS valmAP: 17.82 -> 19.73Improved
Add an others class to every group (ablation)Group softmax with four groups · LVIS valmAP: +2.7%Improved
Add both the background group and others classes (ablation)Group softmax with four groups · LVIS valmAP: 23.74 -> 25.96Improved
Use 2 rather than 4 normal-category groups (ablation)BAGS · LVIS valmAP: 25.96 -> 25.31Hurt
Use 8 rather than 4 normal-category groups (ablation)BAGS · LVIS valmAP: 25.96 -> 24.85Hurt
Increase the others sampling ratio up to β=8 (ablation)BAGS · LVIS valmAPImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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