AI Paper Atlas

Research map / 3D vision / Depth and stereo

Pixel-wise Geo-registration of Drone and Satellite Images

Qingyang Liu, David G Shatwell, Parth Parag Kulkarni, Shah Mubarak

ECCV 2026 · 3D vision › Depth and stereo · approach: Scene coordinate regression for localization · 0 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 RoMa homography warp with 3D-aware SkyReg warpdrone-satellite geo-registration · SkyReg-Bench Urban (pixel-wise)mean geodetic error (m): 133.06 -> 44.88Improved
Replace RoMa homography warp with SkyRegdrone-satellite geo-registration · SkyReg-Bench Suburban (pixel-wise)mean geodetic error (m): 112.10 -> 21.34Improved
Replace RoMa with SkyRegmedian-GPS geo-localization · SkyReg-Bench Urbanmean geodetic error (m): 104.70 -> 36.40Improved
Replace RoMa with SkyRegmedian-GPS geo-localization · SkyReg-Bench Suburbanmean geodetic error (m): 97.78 -> 21.45Improved
Use RoMa dense correspondences instead of sparse matchesRoMa · SkyReg-Bench Urban (pixel-wise)mean geodetic error (m): 133.06 -> 131.45No clear effect
Use RoMa dense correspondences instead of sparse matchesRoMa · SkyReg-Bench Urban (pixel-wise)Recall@20m: 25.69 -> 17.62Hurt
Train on Landmarks subset only instead of Landmarks+Urban (ablation)SkyReg · SkyReg-Bench Urban (median-GPS)mean geodetic error (m): 36.40 -> 48.70Hurt
Train on Urban subset only instead of Landmarks+Urban (ablation)SkyReg · SkyReg-Bench Urban (median-GPS)mean geodetic error (m): 36.40 -> 50.38Hurt
Train on Urban only instead of Landmarks+Urban (ablation)SkyReg · SkyReg-Bench Suburban (median-GPS)mean geodetic error (m): 21.45 -> 24.79Hurt
Strong matcher SP+SG vs cross-view retrieval University-1652baselines · SkyReg-Bench Urban (median-GPS)mean geodetic error (m): 121.17 -> 134.09Hurt

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