OpenMapCD: A Multimodal Benchmark Dataset for Change Detection Between Optical Remote Sensing and Map Data

Overview: 

1. OpenMapCD, the first large-scale multimodal dataset for change detection on optical remote sensing imagery and map (OpenStreetMap) data, supporing basic binary change detection and further semantic change detection
2. OpenMapCD is highly geographically diverse, with 1288 benchmark samples with 1024x1024 pixels from 40 regions across six continents and out-of-distribution data in two areas in Japan
3. Advancing land-cover mapping, binary change detection and semantic change detection tasks, and GIS system updating
   
   

Research Paper: 

• Arxiv paper: https://arxiv.org/html/2310.02674v3
• TGRS paper: https://ieeexplore.ieee.org/document/10551264

Project Page:
The benchmark code is available at: https://github.com/ChenHongruixuan/ObjFormer

Reference:

@ARTICLE{Chen2024ObjFormer,
  author={Chen, Hongruixuan and Lan, Cuiling and Song, Jian and Broni-Bediako, Clifford and Xia, Junshi and Yokoya, Naoto},
  journal={IEEE Transactions on Geoscience and Remote Sensing}, 
  title={ObjFormer: Learning Land-Cover Changes From Paired OSM Data and Optical High-Resolution Imagery via Object-Guided Transformer}, 
  year={2024},
  volume={62},
  number={},
  pages={1-22},
  doi={10.1109/TGRS.2024.3410389}
}