AI Paper Atlas

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

Generative Prompt Model for Weakly Supervised Object Localization

Yuzhong Zhao, Qixiang Ye, Weijia Wu, Chunhua Shen, Fang Wan

ICCV 2023 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised object localization · 35 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
learn representative embedding f_r initialized from f_d instead of using f_d only (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 61.2 -> 64.0Improved
initialize f_r from f_d rather than randomly (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 44.6 -> 64.0Improved
combine f_r and f_d into f_c (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 64.0 -> 64.5Improved
finetune the attention-unet backbone on the WSOL dataset (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 64.5 -> 65.1Improved
aggregate attention maps at timesteps 1 and 100 instead of single timestep (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 64.9 -> 65.1Improved
add large timestep 500 to the aggregation (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 65.1 -> 64.1Hurt
aggregate resolutions 8 and 16 instead of 16 alone (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 62.5 -> 65.1Improved
add 32x32 resolution to the aggregation (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 65.1 -> 64.2Hurt
scale up discriminative TS-CAM with ViT-H backbone and LAION-2B dataTS-CAM · ImageNet-1K valTop-1 Loc: 53.4 -> 42.1Hurt
add noise at inference time (averaged over 10 seeds) (ablation)GenPromp · ImageNet-1K valTop-1 Loc: 64.8 -> 65.1Improved
prompt ensemble over 7 templatesGenPromp · ImageNet-1K valTop-1 Loc: 65.1 -> 65.2Improved

Earlier methods this paper compares with

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