AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Glimpse-Attend-and-Explore: Self-Attention for Active Visual Exploration

Soroush Seifi, Abhishek Jha, Tinne Tuytelaars

ICCV 2021 · Architectures and efficiency › Attention and transformers · approach: Sparse and structured attention · 10 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 contrastive stream (ablation)Glimpse-Attend-and-Explore · SUN360 reconstructionRMSE: 34.2 -> 33.8Improved
add contrastive stream (ablation)Glimpse-Attend-and-Explore · MS-COCO reconstructionRMSE: 43.7 -> 40.3Improved
add contrastive stream (ablation)Glimpse-Attend-and-Explore · ADE20k segmentationmPA (%): 52.3 -> 52.4No clear effect
add contrastive stream (ablation)Glimpse-Attend-and-Explore · COCO-Stuff segmentationmPA (%): 47.5 -> 47.8No clear effect
choose glimpses by self-attention heatmap instead of randomly (ablation)Glimpse-Attend-and-Explore · SUN360 reconstructionRMSE: 39.4 -> 33.8Improved
choose glimpses by self-attention heatmap instead of randomly (ablation)Glimpse-Attend-and-Explore · MS-COCO reconstructionRMSE: 47.8 -> 40.3Improved
choose glimpses by self-attention heatmap instead of randomly (ablation)Glimpse-Attend-and-Explore · ADE20k segmentationmPA (%): 51.5 -> 52.4Improved
replace self-attention with an extra conv channel under random glimpses (ablation)Glimpse-Attend-and-Explore · ADE20k segmentationmPA (%): 51.5 -> 52.4Improved
add attention-based glimpse selection to classifier (ablation)Ours (Full) · SUN360 classificationaccuracy (%): 49.6 -> 56.4Improved
add attention-based glimpse selection to no-decoder classifier (ablation)Ours (No Decoder) · SUN360 classificationaccuracy (%): 62.6 -> 67.2Improved
remove the reconstruction/segmentation decoder for classification (ablation)Glimpse-Attend-and-Explore · SUN360 classificationaccuracy (%): 56.4 -> 67.2Improved
use self-attention glimpse selection instead of Attend and SegmentGlimpse-Attend-and-Explore · ADE20k segmentationmPA: 0.479 -> 0.524Improved

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