RarePlanes

Description

RarePlanes is a unique open-source machine learning dataset from CosmiQ Works and AI.Reverie that incorporates both real and synthetically generated satellite imagery. The RarePlanes dataset specifically focuses on the value of AI.Reverie synthetic data to aid computer vision algorithms in their ability to automatically detect aircraft and their attributes in satellite imagery. Although other synthetic/real combination datasets exist, RarePlanes is the largest openly-available very high resolution dataset built to test the value of synthetic data from an overhead perspective. The real portion of the dataset consists of 253 Maxar WorldView-3 satellite scenes spanning 112 locations and 2,142 km^2 with 14,700 hand-annotated aircraft. The accompanying synthetic dataset is generated via AI.Reverie’s novel simulation platform and features 50,000 synthetic satellite images with ~630,000 aircraft annotations.

Update Frequency

None Planned

License

CC BY-SA 4.0

Documentation

www.cosmiqworks.org/RarePlanes

Managed By

In-Q-Tel - CosmiQ Works

See all datasets managed by In-Q-Tel - CosmiQ Works.

Contact

jss5102@gmail.com and avanetten@iqt.org

How to Cite

RarePlanes was accessed on DATE from https://registry.opendata.aws/rareplanes.

Usage Examples

Tutorials

• Announcing YOLTv4: Improved Satellite Imagery Object Detection by Adam Van Etten
  
• Automatically compress and archive satellite imagery for Amazon S3 by Newel Hirst, Joseph Fahimi, and Justin Downes
  S3S3 GlacierLambdaECR
• Getting Started with YOLTv4 for Object Detection in Imagery: Getting Training Data by Sophia Parafina
  
• Notebook for training and testing YOLYv4 by Adam Van Etten
  

Tools & Applications

• RarePlanes Codebase by Thomas Hossler and Jacob Shermeyer
  

Publications

• RarePlanes: Synthetic Data Takes Flight by Jacob Shermeyer, Thomas Hossler, Adam Van Etten, Daniel Hogan, Ryan Lewis, Daeil Kim