Papers › Joint Representation Learning and Keypoint Detection for Cross-view Geo-localization

Joint Representation Learning and Keypoint Detection for Cross-view Geo-localization

15 May 2022IEEE Transactions on Image Processing (TIP) 2022 5archive 2025-07-28

Jinliang Lin, Zhedong Zheng, Zhun Zhong, Zhiming Luo, Shaozi Li, Yi Yang, Nicu Sebe

In this paper, we study the cross-view geo-localization problem to match images from different viewpoints. The key motivation underpinning this task is to learn a discriminative viewpoint-invariant visual representation. Inspired by the human visual system for mining local patterns, we propose a new framework called RK-Net to jointly learn the discriminative Representation and detect salient Keypoints with a single Network. Specifically, we introduce a Unit Subtraction Attention Module (USAM) that can automatically discover representative keypoints from feature maps and draw attention to the salient regions. USAM contains very few learning parameters but yields significant performance improvement and can be easily plugged into different networks. We demonstrate through extensive experiments that (1) by incorporating USAM, RK-Net facilitates end-to-end joint learning without the prerequisite of extra annotations. Representation learning and keypoint detection are two highly-related tasks. Representation learning aids keypoint detection. Keypoint detection, in turn, enriches the model capability against large appearance changes caused by viewpoint variants. (2) USAM is easy to implement and can be integrated with existing methods, further improving the state-of-the-art performance. We achieve competitive geo-localization accuracy on three challenging datasets, i.e., University-1652, CVUSA and CVACT. Our code is available at https://github.com/AggMan96/RK-Net.

PaperPDFCode

Code

AggMan96/RK-Net mentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Drone navigationDrone-view target localizationImage-Based LocalizationKeypoint DetectionRepresentation Learninggeo-localization

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Drone navigation University-1652 LPN + USAM AP 75.96 #2 of 6 Archive leaderboard report
Drone navigation University-1652 LPN + USAM Recall@1 86.59 #2 of 6 Archive leaderboard report
Drone navigation University-1652 SAFA + USAM AP 71.77 #4 of 6 Archive leaderboard report
Drone navigation University-1652 SAFA + USAM Recall@1 83.23 #4 of 6 Archive leaderboard report
Drone navigation University-1652 RK-Net AP 65.76 #5 of 6 Archive leaderboard report
Drone navigation University-1652 RK-Net Recall@1 80.17 #5 of 6 Archive leaderboard report
Drone-view target localization University-1652 LPN + USAM AP 80.55 #6 of 11 Archive leaderboard report
Drone-view target localization University-1652 LPN + USAM Recall@1 77.60 #6 of 11 Archive leaderboard report
Drone-view target localization University-1652 SAFA + USAM AP 75.79 #8 of 11 Archive leaderboard report
Drone-view target localization University-1652 SAFA + USAM Recall@1 72.19 #8 of 11 Archive leaderboard report
Drone-view target localization University-1652 RK-Net AP 70.23 #10 of 11 Archive leaderboard report
Drone-view target localization University-1652 RK-Net Recall@1 66.13 #10 of 11 Archive leaderboard report
Image-Based Localization cvact RK-Net Recall@1 40.53 #7 of 8 Archive leaderboard report
Image-Based Localization cvact RK-Net Recall@1 (%) 89.12 #7 of 8 Archive leaderboard report
Image-Based Localization cvact Instance Loss Recall@1 35.24 #8 of 8 Archive leaderboard report
Image-Based Localization cvact Instance Loss Recall@1 (%) 87.34 #8 of 8 Archive leaderboard report
Image-Based Localization cvusa RK-Net Recall@1 52.50 #7 of 8 Archive leaderboard report
Image-Based Localization cvusa RK-Net Recall@top1% 96.52 #7 of 8 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections