Research map / Detection and segmentation / 3D detection and driving
Recurrent Cross-View Object Geo-Localization
Xiaohan Zhang, Siyuan Cao, Xiaokai Bai, Yiming Li, Zhangkai Shen, Zhe Wu, Lun Luo, Qi Ming, Xiaoxi Hu, Hui-Liang Shen
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add recurrent refinement (m=5 vs m=1) (ablation) | ReCOT · CVOGL Ground->Satellite (val) | Acc@0.50: 42.25 -> 43.66 | Improved |
| add recurrent refinement (m=5 vs m=1) (ablation) | ReCOT · CVOGL Drone->Satellite (val) | Acc@0.50: 67.06 -> 67.82 | Improved |
| add recurrent refinement (m=5 vs m=1) (ablation) | ReCOT · CVOGL Ground->Satellite (test) | Acc@0.50: 46.25 -> 48.10 | Improved |
| increase steps beyond m=5 to m=6 (ablation) | ReCOT · CVOGL Ground->Satellite (val) | Acc@0.50: 43.66 -> 43.34 | No clear effect |
| add recurrent steps in few-shot (m=1 to m=6) (ablation) | ReCOT · CVOGL-few-shot Drone->Satellite (test) | Acc@0.50: 25.00 -> 33.33 | Improved |
| add RFEM (ablation) | ReCOT · CVOGL Ground->Satellite (test) | Acc@0.50: 42.24 -> 48.10 | Improved |
| add RFEM (ablation) | ReCOT · CVOGL Drone->Satellite (test) | Acc@0.50: 46.04 -> 72.05 | Improved |
| add SAM distillation loss (ablation) | ReCOT · CVOGL Drone->Satellite (test) | Acc@0.50: 65.78 -> 72.05 | Improved |
| add token guidance loss (ablation) | ReCOT · CVOGL Ground->Satellite (test) | Acc@0.50: 47.58 -> 48.10 | Improved |
| add spatial attention map M in RFEM (ablation) | ReCOT · CVOGL Drone->Satellite (test) | Acc@0.50: 70.30 -> 72.05 | Improved |
| use low-res for attention and high-res for enhancement in RFEM (ablation) | ReCOT · CVOGL Ground->Satellite (test) | Acc@0.50: 43.28 -> 48.10 | Improved |
| use 100 tokens instead of 1 (ablation) | ReCOT · CVOGL Ground->Satellite (test) | Acc@0.50: 43.37 -> 48.10 | Improved |
Closest papers by meaning
- Breaking Rectangular Shackles: Cross-View Object Segmentation for Fine-Grained Object Geo-Localization (ICCV 2025 · similarity 0.91)
- MOGeo: Beyond One-to-One Cross-View Object Geo-localization (CVPR 2026 · similarity 0.90)
- SinGeo: Unlock Single Model's Potential for Robust Cross-View Geo-Localization (CVPR 2026 · similarity 0.89)
- Aligning Geometric Spatial Layout in Cross-View Geo-Localization via Feature Recombination (AAAI 2024 · similarity 0.89)
- Unleashing Unlabeled Data: A Paradigm for Cross-View Geo-Localization (CVPR 2024 · similarity 0.88)
- From Coarse to Fine: A Matching and Alignment Framework for Unsupervised Cross-View Geo-Localization (AAAI 2025 · similarity 0.88)
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.
