AI Paper Atlas

Research map / 3D vision / Depth and stereo

MAD-DR: Map Compression for Visual Localization with Matchness Aware Descriptor Dimension Reduction

Qiang Wang

ECCV 2024 · 3D vision › Depth and stereo · approach: Scene coordinate regression for localization · 5 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
jointly retrain the matcher with compressed descriptors instead of using pre-trained SuperGlue (ablation)MAD-DR (SuperPoint, D_LR=8) · Aachen Night (0.25m,2deg)localization accuracy (%): 71.4 -> 82.7Improved
use asymmetric design (uncompressed query, compressed database) instead of symmetric training (ablation)MAD-DR · InLoc duc1 / duc2localization accuracy (%): duc1 30.8 -> 46.5; duc2 29.0 -> 48.1 (vs Sym db only)Improved
add descriptor reconstruction loss (ablation)MAD-DR · InLoc duc1localization accuracy (%): 40.9 -> 46.5Improved
add descriptor reconstruction loss (ablation)MAD-DR · Aachen Nightlocalization accuracy (%): 83.7 -> 82.7No clear effect
apply encoder-decoder to both query and database after asymmetric training (ablation)MAD-DR · Aachen Nightlocalization accuracy (%): 82.7 -> 83.7No clear effect
compress SuperPoint descriptors to dimension 4 (1/256 size) with joint trainingSuperPoint+SuperGlue · InLoc duc1 (0.25m,2deg)localization accuracy (%): 49.5 -> 48.0No clear effect
joint training compression versus SceneSqueezer on Aachen nightSceneSqueezer · Aachen Night (0.25m,2deg)localization accuracy (%): 50.0 -> 80.6 at 31 MB -> 4.1 MBImproved
replace DISK+MNN+PQ baseline with joint-trained compression on InLocDISK+MNN+PQ · InLoclocalization accuracy (%): DISK+MNN does not perform well (>15% worse) on InLocImproved
apply method to DISK instead of product quantizationDISK · Aachen Nightlocalization accuracy (%): 70.4 -> 84.7 with 1/4 memoryImproved

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