https://github.com/xecades/aerialextrematch

(WIP) AerialExtreMatch: A Benchmark for Extreme-View Image Matching and Localization
https://github.com/xecades/aerialextrematch

benchmark cv dataset image-matching localization

Last synced: 5 months ago
JSON representation

(WIP) AerialExtreMatch: A Benchmark for Extreme-View Image Matching and Localization

• Host: GitHub
• URL: https://github.com/xecades/aerialextrematch
• Owner: Xecades
• License: mit
• Created: 2025-05-05T09:48:13.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2025-05-24T07:15:17.000Z (about 1 year ago)
• Last Synced: 2025-05-24T08:19:18.245Z (about 1 year ago)
• Topics: benchmark, cv, dataset, image-matching, localization
• Homepage: https://xecades.github.io/AerialExtreMatch/
• Size: 98.1 MB
• Stars: 5
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# AerialExtreMatch: A Benchmark for Extreme-View Image Matching and Localization

### [Project Page](https://xecades.github.io/AerialExtreMatch/) | Paper (WIP)




> **AerialExtreMatch: A Benchmark for Extreme-View Image Matching and Localization**  

> [Rouwan Wu1](https://github.com/RingoWRW), [Zhe Huang2](https://github.com/Xecades), [Xingyi He2](https://hxy-123.github.io/), [Yan Liu3](https://faculty.hdu.edu.cn/jsjxy/ly2_21682/main.htm), [Shen Yan1](https://openreview.net/profile?id=~Shen_Yan6), [Sida Peng2](https://pengsida.net/), [Maojun Zhang1†](https://orcid.org/0000-0001-6748-0545), [Xiaowei Zhou2†](https://xzhou.me/)  

> 1NUDT, 2State Key Lab of CAD&CG, ZJU, 3HUST  

>  2025



    

    


    We introduce AerialExtreMatch, a large-scale, high-fidelity benchmark tailored for extreme-view image matching and UAV localization. It consists of three datasets: Train Pair, Evaluation Pair, and Localization. All code and datasets are readily available for public access.



> [!WARNING]  

> Docs are under preparation, and will be released soon.

## Resources

> [!IMPORTANT]  

> In our paper, TWO seperate codebases are provided: **benchmarking** and code of our pretrained **RoMa** model.  

> To increase simplicity and consistency, we slightly abuse the concept of git branches and **make the two codebases as branches of this repository**.

 - **Code**

   - [**`Benchmark` branch**](https://github.com/Xecades/AerialExtreMatch/tree/Benchmark): source code for the benchmark, including feature matching and localization pipelines for models mentioned in the paper. **(WIP)**

   - [**`RoMa` branch**](https://github.com/Xecades/AerialExtreMatch/tree/RoMa): the code we use to train our RoMa model.

 - **Dataset**

   - [**AerialExtreMatch-Train**](https://huggingface.co/datasets/Xecades/AerialExtreMatch-Train): corresponds to **Train Pair** set.

   - [**AerialExtreMatch-Benchmark**](https://huggingface.co/datasets/Xecades/AerialExtreMatch-Benchmark): corresponds to **Evaluation Pair** set.

   - [**AerialExtreMatch-Localization**](https://huggingface.co/datasets/Xecades/AerialExtreMatch-Localization): corresponds to **Localization** set.

 - **Checkpoints**: see [[Release]](releases).

## License

[MIT License](LICENSE)