AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

AdaGlimpse: Active Visual Exploration with Arbitrary Glimpse Position and Scale

Adam Pardyl, Michał Wronka, Maciej Wołczyk, Kamil Adamczewski, Tomasz Trzcinski, Bartosz Zieliński

ECCV 2024 · Reinforcement learning › Visual and representation RL · approach: Robust visual encoders for RL · 7 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
select glimpses at arbitrary position and scale instead of fixed-scale gridsAdaGlimpse · ImageNet-1k classificationaccuracy (%): 76.13 -> 77.54Improved
add early stopping based on predicted class probabilityAdaGlimpse · ImageNet-1k classificationpixel percentage (%): 28.57 -> ~16.94Improved
use adaptive scale glimpses instead of simple grids at same pixel budgetAdaGlimpse · ImageNet-1k reconstruction (6% pixels)RMSE: 20.9 vs 23.4 (AME, 18.75% pixels, SUN360 best baseline)Improved
adaptive glimpses versus AME at similar pixel budgetAdaGlimpse · ADE20K segmentationPA (%): 70.3 -> 70.0No clear effect
replace glimpse coordinates in RL state with dataset mean (ablation)AdaGlimpse RL agent · ImageNet-1k classificationaccuracy (%): 68.25 -> 77.54Improved
include transformer latent in RL state (ablation)AdaGlimpse RL agent · ImageNet-1k classificationaccuracy (%): 61.82 -> 77.54Improved
include patch pixels in RL state (ablation)AdaGlimpse RL agent · ImageNet-1k classificationaccuracy (%): 76.99 -> 77.54Improved
include patch importance in RL state (ablation)AdaGlimpse RL agent · ImageNet-1k classificationaccuracy (%): 77.36 -> 77.54Improved
train only on ImageNet-1k and test zero-shot on other reconstruction datasetsAdaGlimpse · SUN360/ADE20K/COCO reconstructionRMSE: 11.1 vs 23.6 (SUN360, AME)Improved

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