AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

C-WSL: Count-guided Weakly Supervised Localization

Mingfei Gao, Ang Li, Ruichi Yu, Vlad I. Morariu, Larry S. Davis

ECCV 2018 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised object localization · 96 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 alternating detector refinement (ADR) to WSLPDA (ablation)WSLPDA · VOC2007 testmAP: 39.5 -> 42.7Improved
add count-based region selection (CRS) to WSLPDA+ADR (ablation)WSLPDA+ADR · VOC2007 testmAP: 42.7 -> 45.7Improved
add alternating detector refinement (ADR) to OICR (ablation)OICR · VOC2007 testmAP: 41.2 -> 44.9Improved
add count-based region selection (CRS) to OICR+ADR (ablation)OICR+ADR · VOC2007 testmAP: 44.9 -> 46.4Improved
add count supervision to WSLPDA+ADR (ablation)WSLPDA+ADR · VOC2007 trainvalCorLoc: 59.2 -> 61.9Improved
add count supervision to OICR+ADR (ablation)OICR+ADR · VOC2007 trainvalCorLoc: 63.5 -> 64.6Improved
add count supervision to WSLPDA+ADR (ablation)WSLPDA+ADR · VOC2012 valmAP: 40.0 -> 41.3Improved
add count supervision to OICR+ADR (ablation)OICR+ADR · VOC2012 valmAP: 39.6 -> 41.5Improved
add count supervision to WSLPDA+ADR (ablation)WSLPDA+ADR · VOC2012 trainCorLoc: 61.9 -> 62.0No clear effect
add count supervision to sheep class of WSLPDA+ADR (ablation)WSLPDA+ADR · VOC2007 (sheep class)CorLoc: 13.4% CorLoc and 10.4% APImproved
add count supervision to sheep class of OICR+ADR (ablation)OICR+ADR · VOC2007 (sheep class)CorLoc: 3.1% CorLoc and 6.1% APImproved
apply C-WSL to OICR-Ens.+FRCNN on COCO (20 common classes, no fine-tuning)OICR-Ens.+FRCNN · MS COCO 35k-val2014 (20 classes)mAP@0.5: 19.0 -> 20.0Improved

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