AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Geometrized Transformer for Self-Supervised Homography Estimation

Jiazhen Liu, Xirong Li

ICCV 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Pretext-task visual pre-training · 37 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 RANSAC-based homography to select updated query points (ablation)GeoFormer · FIREmAUC: 67.3 -> 75.6Improved
add RANSAC-based homography (ablation)GeoFormer · HPatchesmAUC: 71.4 -> 76.7Improved
add focused cross-attention (ablation)GeoFormer · HPatchesmAUC: 73.0 -> 76.7Improved
add sparse self-attention (ablation)GeoFormer · HPatchesmAUC: 75.4 -> 76.7Improved
use standard instead of linear Transformers (ablation)GeoFormer · HPatchesmAUC: 73.4 -> 76.7Improved
use fine-matching2 (symmetric) instead of LoFTR fine-matching (ablation)GeoFormer · ISC-HEmAUC: 41.8 -> 44.0Improved
add GeoFormer to LoFTRLoFTR · HPatchesAUC@3px vs ASpanFormer: 59.9 -> 68.0 (vs ASpanFormer)Improved
replace LoFTR with GeoFormer for visual localizationLoFTR · Aachen Day-Night (0.25m,2deg)accuracy: 79.6 -> 77.6Hurt
add GeoFormer to LoFTRLoFTR · 640x480 image pair inferenceruntime (ms): 77 -> 101Hurt

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