AI Paper Atlas

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

Evaluation for Weakly Supervised Object Localization: Protocol, Metrics, and Datasets

Junsuk Choe, Seong Joon Oh, Sanghyuk Chun, Seungho Lee, Zeynep Akata, Hyunjung Shim

TPAMI 2022 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised object localization · 28 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 HaS erasing augmentationCAM · Total mean (3 datasets)mean localization: -1.4Hurt
add ACoL adversarial erasing headCAM · Total mean (3 datasets)mean localization: -3.7Hurt
add SPG self-produced guidanceCAM · Total mean (3 datasets)mean localization: -0.9Hurt
add ADL attention-based dropoutCAM · Total mean (3 datasets)mean localization: -1.8Hurt
add CutMix augmentationCAM · Total mean (3 datasets)mean localization: -0.5Hurt
replace CAM with GC-NetCAM · ImageNetMaxBoxAccV2: -9.9Hurt
use few-shot fully supervised training instead of WSOL hyperparameter searchCAM / best WSOL · Total mean (3 datasets)mean localization: 64.5 -> 72.3Improved
use max normalization instead of min-max for score maps (ablation)WSOL methods · Total mean (3 datasets)mean localization: CutMix +0.2, ACoL -5.4No clear effect
use CAM score maps instead of input attribution (VB)input attribution methods · OpenImagesmPxAP: 54.8 -> 60.5Improved
add SmoothGrad to vanilla backprop (ablation)VB · OpenImagesPxAP: 54.0 to 49.6Hurt
add Integrated Gradients to vanilla backprop (ablation)VB · OpenImagesPxAP: 54.0 to 57.1Improved
add Integrated Gradients to guided backprop (ablation)GB · OpenImagesPxAP: 53.4 to 53.9No clear effect

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