AI Paper Atlas

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

Weakly Supervised Object Localization via Transformer with Implicit Spatial Calibration

Haotian Bai, Ruimao Zhang, Jiong Wang, Xiang Wan

ECCV 2022 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised object localization · 45 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 SCM (dropped at inference) to TS-CAMTS-CAM (Deit-S) · CUB-200-2011GT-Known: 87.7 -> 96.6Improved
add SCM (dropped at inference) to TS-CAMTS-CAM (Deit-S) · CUB-200-2011Loc Acc Top-1: 71.3 -> 76.4Improved
add SCM (dropped at inference) to TS-CAMTS-CAM (Deit-S) · CUB-200-2011Cls Acc Top-1: 80.3 -> 78.5Hurt
add SCM to TS-CAMTS-CAM (Deit-S) · ILSVRC2012 valGT-Known: 67.6 -> 68.8Improved
add SCM to TS-CAMTS-CAM (Deit-S) · ILSVRC2012 valLoc Acc Top-1: 53.4 -> 56.1Improved
add SCM to TS-CAMTS-CAM (Deit-S) · CUBV2 (held-out)MaxboxAccV2: 79.6 -> 89.9Improved
add SCM to TS-CAMTS-CAM (Deit-S) · CUBV2 (held-out)MaxboxAccV1: 88.9 -> 96.6Improved
use SCM with ViT backboneViT · CUB-200-2011GT-Known: 91.8Improved
use SCM with Conformer backboneConformer · CUB-200-2011GT-Known: 96.1Improved
stack more activation diffusion blocks (ablation)SCM · CUB-200-2011GT-KnownMixed
use larger Deit backbones instead of Deit-small (ablation)SCM · CUB-200-2011GT-KnownHurt
use Newton-Schulz iteration p=4 instead of other iteration counts (ablation)SCM · CUB-200-2011GT-KnownImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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