AI Paper Atlas

Research map / Detection and segmentation / CNN features

S-TREK: Sequential Translation and Rotation Equivariant Keypoints for Local Feature Extraction

Emanuele Santellani, Christian Sormann, Mattia Rossi, Andreas Kuhn, Friedrich Fraundorfer

ICCV 2023 · Detection and segmentation › CNN features · approach: Local features and context in CNNs · 15 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
replace peaky+similarity heatmap losses with sequential-sampling RL-style training (ablation)ReCNN detector · Synthetic lines validationmax repeatability@1px: 0.58 -> 0.78Improved
vary number of sequential samples (50, 100, 200) (ablation)S-TREK detector · Synthetic lines validationnumber of detected keypoints: 39, 53, 56 keypointsNo clear effect
vary peaky window size (32, 64, 96) (ablation)peaky+similarity baseline · Synthetic lines validationnumber of detected keypoints: 27 to 60Hurt
use rotation-equivariant convolutions instead of standard CNN with same channels (ablation)S-TREK detector · Synthetic lines validationmax repeatability@1px: 0.61 -> 0.78Improved
use ReCNN instead of a much larger standard CNN (377k params) (ablation)S-TREK detector · Synthetic lines validationmax repeatability@1px: 0.68 -> 0.78Improved
use a smaller shallow ReCNN (cyclic 8) detector (ablation)S-TREK detector · HPatches rotation repeatabilitymean repeatabilityHurt
increase cyclicity from 8 to 16 in the smaller ReCNN (ablation)smaller ReCNN · HPatches rotation repeatabilitymean repeatabilityNo clear effect
rotation-equivariant detector with detect-then-describeS-TREK · HPatches +-45 degHomography accuracy AUC@3px: 0.182 -> 0.248Improved
S-TREK vs DISK on unrotated Image Matching BenchmarkS-TREK · Image Matching Benchmark standardmAA@10: 0.512 vs 0.458Hurt

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