BRIGHT: A globally distributed multimodal building damage dataset
Multimodal building damage imagery across 7 disaster types in 12 global regions
- Source
- Zenodo
- Published
- 2025
- Format
- multiple formats
- Access
- Open access
- License
- CC-BY-4.0
- Themes
- Natural Disasters, Machine Learning & AI, Remote Sensing & Satellites
About this dataset
BRIGHT is the first open-access, globally distributed, event-diverse multimodal dataset specifically curated to support AI-based disaster response. It covers five types of natural disasters and two types of man-made disasters across 12 regions worldwide, with a particular focus on developing countries.
How the data was collected
Data collected from various global events, focusing on building damage assessment through multimodal sources, including satellite imagery and ground reports. The dataset is structured to aid in AI-based disaster response modeling.