Hongruixuan Chen1,2, Jian Song1,2, Olivier Dietrich3, Clifford Broni-Bediako2, Weihao Xuan1,2, Junjue Wang1
Xinlei Shao1, Yimin Wei1,2, Junshi Xia3, Cuiling Lan4, Konrad Schindler3, Naoto Yokoya1,2 *
1 The University of Tokyo, 2 RIKEN AIP, 3 ETH Zurich, 4 Microsoft Research Asia
Notice☀️☀️: The full version of the BRIGHT paper is now online.Nov 18th, 2025: BRIGHT has been accepted by ESSD and online available now!!Aug 12th, 2025: You can check winners' solutions by visiting their Github repos: 1st rank and 2nd rank!!Apr 28th, 2025: IEEE GRSS DFC 2025 Track II is over. Congratulations to winners!! You can now download the full version of DFC 2025 Track II data in Zenodo or HuggingFace!!Jan 13th, 2025: The benchmark code for IEEE GRSS DFC 2025 Track II is now available. Please follow the instruction to use it!! Also, you can find dataset and code related to Track I in here!!
-
BRIGHT also serves the official dataset of 2025 IEEE GRSS Data Fusion Contest: All-Weather Land Cover and Building Damage Mapping.
-
Although the DFC25 has concluded, we have retained the relevant code and dataset splits to support your research if you wish to continue working within the DFC25 setting. Specifically, nine events are designated for training and validation, while two events (Noto-Earthquake-2024 and Marshall-Wildfire-2021) are reserved unseen for testing.
Note that the code in this repo runs under Linux system. We have not tested whether it works under other OS.
Step 1: Clone the repository:
Clone this repository and navigate to the project directory:
git clone https://github.com/ChenHongruixuan/BRIGHT.git
cd BRIGHTStep 2: Environment Setup:
It is recommended to set up a conda environment and installing dependencies via pip. Use the following commands to set up your environment:
Create and activate a new conda environment
conda create -n bright-benchmark
conda activate bright-benchmarkInstall dependencies
pip install -r requirements.txtPlease download the BRIGHT from Zenodo or HuggingFace. Note that you don't need data over Ukraine, Myanmar, and Mexico under DFC 25 setup.
After the data has been prepared, please make them have the following folder/file structure:
${DATASET_ROOT} # Dataset root directory, for example: /home/username/data/bright
│
├── pre-event
│ ├──bata-explosion_00000000_pre_disaster.tif
│ ├──bata-explosion_00000001_pre_disaster.tif
│ ├──bata-explosion_00000002_pre_disaster.tif
│ ...
│
├── post-event
│ ├──bata-explosion_00000000_post_disaster.tif
│ ...
│
└── target
├──bata-explosion_00000000_building_damage.tif
...
The following commands show how to train and evaluate UNet on the BRIGHT dataset using DFC25 split set in [bda_benchmark/dataset/splitname/dfc25]:
python script/standard_ML/train_UNet.py --dataset 'BRIGHT' \
--train_batch_size 16 \
--eval_batch_size 4 \
--num_workers 16 \
--crop_size 640 \
--max_iters 800000 \
--learning_rate 1e-4 \
--model_type 'UNet' \
--model_param_path '<your model checkpoint saved path>' \
--train_dataset_path '<your dataset path>' \
--train_data_list_path '<your project path>/bda_benchmark/dataset/splitname/dfc25/train_set.txt' \
--val_dataset_path '<your dataset path>' \
--val_data_list_path '<your project path>/bda_benchmark/dataset/splitname/dfc25/holdout_set.txt' \
--test_dataset_path '<your dataset path>' \
--test_data_list_path '<your project path>/bda_benchmark/dataset/splitname/dfc25/val_set.txt' The following commands show how to evaluate trained UNet on the two unseen events:
python script/standard_ML/infer_UNet.py --model_path '<path of the checkpoint of model>' \
--test_dataset_path '<your dataset path>' \
--test_data_list_path '<your project path>/bda_benchmark/dataset/splitname/dfc25/test_set.txt' \
--output_dir '<your inference results saved path>'Then, you can still go to the official Leaderboard in CodaLab to submit your results.
Keep the prediction name consistent with label name, i.e., turkey-earthquake_00000001_building_damage.png, hawaii-wildfire_00000003_building_damage.png, and so on.All png files should be submitted in zip file format. Zip all prediction files directly without any folders containing them.Using the raw prediction results instead of visualized ones.
Based on peers' questions from issue section, here's a quick navigate list of solutions to some common issues.
| Issue | Solution |
|---|---|
| Abnormal accuracy (like 0 or -999999) given by leaderboard | Keep the prediction name and label name consistent / Zip all prediction files directly, not the folder containing them |
| Leaderboard server not responding after submitting results | Change browser (Google Chrome recommended) |
| Python package conflicts | The baseline code is not limited to a specific version, and participants do not need to match the version we provide. |
If this dataset or code contributes to your research, please kindly consider citing our paper and give this repo ⭐️ :)
@Article{Chen2025Bright,
AUTHOR = {Chen, H. and Song, J. and Dietrich, O. and Broni-Bediako, C. and Xuan, W. and Wang, J. and Shao, X. and Wei, Y. and Xia, J. and Lan, C. and Schindler, K. and Yokoya, N.},
TITLE = {\textsc{Bright}: a globally distributed multimodal building damage assessment dataset with very-high-resolution for all-weather disaster response},
JOURNAL = {Earth System Science Data},
VOLUME = {17},
YEAR = {2025},
NUMBER = {11},
PAGES = {6217--6253},
DOI = {10.5194/essd-17-6217-2025}
}
The authors would also like to give special thanks to Sarah Preston of Capella Space, Capella Space's Open Data Gallery, Maxar Open Data Program and Umbra Space's Open Data Program for providing the valuable data.
For any questions, please feel free to leave it in the issue section or contact us.