The Intelligence Advanced Research Projects Activity, within the Office of the Director of National Intelligence, announced on July 18 the launch of the functional Map of the World —“fMoW”— Challenge. The challenge invites solvers from around the world to develop machine learning algorithms and other automated techniques that accurately detect and categorize classify points of interest from satellite imagery.
The goal of the fMoW Challenge is to promote and benchmark research in object recognition and categorization from satellite imagery for automated identification of facility, building, and land use. A satellite imagery dataset collected commercially with one million points of interest annotated will be provided for researchers and entrepreneurs, enabling them to better their methods for understanding satellite imagery through novel learning frameworks and multi-modal fusion techniques.
“Although deep learning has been making a really big impact in many areas including image processing, it has not been applied to the satellite imagery domain extensively,” said Dr. HakJae Kim, IARPA program manager. “Going into the challenge, these will be the largest functionally annotated databases of satellite imagery made available to the public, and we are excited to see what outcomes will be revealed.”
IARPA invites experts from across academia, industry, and developer communities—with or without experience in satellite image analysis—to participate in a convenient, efficient, and non-contractual way. IARPA will provide solvers with a predetermined point-of-interest category library and image sets containing numerous unidentified points as training data. Participants will be asked to generate an algorithm to detect and categorize building and land use in the provided images.
Participants will generate algorithms to detect and categorize facility, building, and land use in the provided satellite images. Throughout the challenge, an online leaderboard will display solvers’ rankings and accomplishments, giving them various opportunities to have their work viewed and appreciated by stakeholders from industry, government, and academic communities. Solvers who are eligible to win a prize and with the most accurate and complete solutions will be eligible to win cash prizes from a total prize purse of $100,000.
To learn more about the functional Map of the World Challenge, including rules, criteria and eligibility requirements, visit www.iarpa.gov/challenges/fmow.html.