AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Adaptive Spot-Guided Transformer for Consistent Local Feature Matching

Jiahuan Yu, Jiahao Chang, Jianfeng He, Tianzhu Zhang, Jiyang Yu, Feng Wu

CVPR 2023 · Architectures and efficiency › Attention and transformers · approach: Efficient image transformers · 34 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 multi-level cross attention at 1/32 resolution (ablation)ASTR baseline · MegaDepthpose estimation AUC@5°: 45.6 -> 46.7Improved
add spot-guided attention at 1/8 resolution in place of linear attention (ablation)ASTR with multi-level cross attention · MegaDepthpose estimation AUC@5°: 46.7 -> 47.7Improved
add adaptive scaling (ablation)ASTR with multi-level and spot-guided attention · MegaDepthpose estimation AUC@5°: 47.7 -> 48.3Improved
vary k from 1 to 6 with l=5 (ablation)ASTR spot-guided attention · MegaDepthpose estimation AUC@5°: 46.0; 47.5; 47.3; 47.7; 47.1; 46.9Mixed
vary l from 3 to 9 with k=4 (ablation)ASTR spot-guided attention · MegaDepthpose estimation AUC@5°: 46.7; 47.7; 47.2; 43.0Mixed
use ASTR instead of PDC-Net+HPatches homography estimation · HPatcheshomography estimation AUC: +4.4% at 3px; +3.5% at 5px; +2.5% at 10pxImproved
use ASTR instead of ASpanFormerMegaDepth relative pose estimation · MegaDepthpose estimation AUC: +3.1% at 5°; +1.6% at 10°Improved
use ASTR instead of QuadtreeScanNet relative pose estimation (models not trained on ScanNet) · ScanNetpose estimation AUC: 19.0 -> 19.4; 37.3 -> 37.6; 53.5 -> 54.4Improved
use ASTR instead of ASpanFormerInLoc visual localization · InLoc DUC1registered images (%): 51.5 -> 53.0; 73.7 -> 73.7; 86.4 -> 87.4Mixed
use ASTR instead of ASpanFormerAachen visual localization · Aachen Day and Nightregistered images (%): 89.4 -> 89.9; 95.6 -> 95.6; 99.0 -> 99.2; 77.5 -> 76.4; 91.6 -> 92.1; 99.0 -> 99.5Mixed

Earlier methods this paper compares with

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