Nick Weir, a data scientist at In-Q-Tel and challenge manager for SpaceNet project, told Federal News Network in an interview published Monday how IQT's CosmiQ Works laboratory leverages artificial intelligence through the SpaceNet initiative.
Weir noted that IQT collaborates with Maxar Technologies, Amazon Web Services and Intel AI Lab to advance the development of algorithms designed to identify building footprints using datasets of imagery through SpaceNet coding challenges run with Topcoder.
He mentioned the potential role of computer vision in accelerating the mapping process following a natural disaster and explained how SpaceNet uses crowdsourcing.
“We get usually between 50 and 250 competitors participating in these challenges, and everyone is trying different algorithms to try to solve this problem,“ Weir said. “Everyone has their own machine learning solution to try to find buildings, and this allows us to figure out which solution works best.“