NGA wants support for its 3D point cloud extraction efforts

3D point cloud The National Geospatial-Intelligence Agency (NGA) issued a “sources sought” notice on July 30 looking for a vendor that could support its InnoVision Basic and Applied Research Office by automatically deriving “dense, accurate, and georeferenced three dimensional point clouds from commercial satellite imagery.”

The agency’s RFI seeks to “discover the state of the art in 3D point cloud extraction from multiple 2D commercial satellite images in order to determine candidates for a future characterization of mature capabilities,” explains the notice.

The work will be performed in Springfield, VA.

Prospective vendors are asked to describe how they could:

  • Produce spatially accurate, high fidelity, dense 3D point clouds.
    • Accommodate typical Earth scene types including urban, rural, feature rich, and feature poor.
    • Accommodate natural changes in solar illumination caused by seasonal and diurnal fluctuations.
    • Produce point cloud point spacing equal to input imagery resolution.
    • Ingest many overlapping images (3 to 100s) over the same geographic location.
    • Process images covering large areas (10s to 100s square km).
    • Accept images with small and large angular diversity.

Further information is available from Kebra Beckford, a contract specialist, at 571-557-2430 or kebra.l.beckford@nga.mil