AI Paper Atlas

Research map / Point clouds / Point cloud classification

MMGeo: Multimodal Compositional Geo-Localization for UAVs

Yuxiang Ji, Boyong He, Zhuoyue Tan, Liaoni Wu

ICCV 2025 · Point clouds › Point cloud classification · approach: LiDAR point cloud compression and completion · 12 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 text modality to the query (I+T->I)Game4Loc · GTA-UAV-MM cross-areaR@1: 52.03% -> 55.34%Improved
add point cloud to the query (I+P->I)Game4Loc · GTA-UAV-MM cross-areaR@1: 52.03% -> 54.75%Improved
add depth to the query (I+D->I)Game4Loc · GTA-UAV-MM cross-areaDis@1: 365.65m -> 297.45mImproved
add text modality to the queryGame4Loc · UAV-VisLoc-MM cross-areaR@1: 47.56% -> 53.47%Improved
add point cloud to the queryGame4Loc · UAV-VisLoc-MM same-areaR@1: 89.86% -> 94.20%Improved
train with text but test with image only (ablation)MMGeo · GTA-UAV-MM cross-areaR@1: 52.03% -> 53.11%Improved
train with point cloud but test with image only (ablation)MMGeo · GTA-UAV-MM cross-areaR@1: 52.03% -> 53.27%Improved
add text to image query under partial occlusion (ablation)MMGeo · GTA-UAV-MM cross-areaR@1: 28.78% -> 41.03%Improved
add text to image query under pixelation (ablation)MMGeo · GTA-UAV-MM cross-areaR@1: 7.61% -> 13.64%Improved
add text to image query under salt and pepper noise (ablation)MMGeo · GTA-UAV-MM cross-areaR@1: 29.12% -> 33.30%Improved
add learnable [SUB] substitution token (length 50) for text (ablation)MMGeo I+T · GTA-UAV-MM cross-areaR@1: 52.88% -> 55.34%Improved
add learnable [SUB] substitution token (length 500) for point cloud (ablation)MMGeo I+P · GTA-UAV-MM cross-areaR@1: 53.71% -> 54.75%Improved

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