PP2-M: Multimodal Geospatial Data for Geo-Foundation Models

PP2-M is a multimodal dataset derived from Place Pulse 2.0, enriched with additional geospatial modalities. The dataset includes aligned pairs of street view images, remote sensing images from Sentinel-2, cartographic data, and geographical coordinates. It was created by author DominikM198 and last updated on HuggingFace in May 2026.

Use Cases

• Train multimodal Geo-Foundation Models based on aligned pairs of street view and satellite imagery.
• Analyze urban environments based on combined geographical coordinates and remote sensing data.
• Develop computer vision models for geospatial understanding based on cartographic and image data.

Strengths

• Dataset is explicitly designed for training multimodal Geo-Foundation Models.
• Includes aligned pairs of multiple geospatial modalities.
• Based on the established Place Pulse 2.0 dataset.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Derived from Place Pulse 2.0 and Sentinel-2.

Collection Method

Enriched from original Place Pulse 2.0 dataset.

Freshness

Last updated 2026-05-29 21:26:54; freshness should be verified.