BRIGHT: Multimodal Satellite Imagery for Disaster Response Across 14 Regions

An open-access multimodal dataset curated by Kullervo for AI-based disaster response. It contains about 4,200 paired optical and SAR images covering five natural and two man-made disaster types across 14 global regions, with a focus on developing countries. The dataset includes over 380,000 building instances at spatial resolutions between 0.3 and 1 meter.

Use Cases

• Building damage mapping based on over 380,000 annotated building instances.
• Zero-shot classification of disaster types based on the seven covered disaster categories.
• Image segmentation of affected areas based on high-resolution optical and SAR imagery.
• Feature extraction for disaster response models based on paired multimodal satellite data.

Strengths

• Contains about 4,200 paired optical and SAR images, providing multimodal data.
• Covers over 380,000 building instances for detailed analysis.
• Includes data from 14 regions worldwide with a focus on developing countries.
• Images have a spatial resolution between 0.3 and 1 meter, offering detailed representations.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• License information is unknown, which may restrict usage.

Provenance

Source

Kullervo on Hugging Face.

Collection Method

Curated specifically to support AI-based disaster response.

Time Range

null

Freshness

Last updated 2026-03-25 06:33:40; freshness should be verified.

Geography

Covers 14 regions worldwide, with a particular focus on developing countries.