InThe DownLinQbyDaniel Hogan·Feb 8, 2021A Closer Look at the SpaceNet Change and Object Tracking (SCOT) MetricPreface: SpaceNet LLC is a nonprofit organization dedicated to accelerating open source, artificial intelligence applied research for…
InThe DownLinQbyDaniel Hogan·Sep 25, 2020Deploying the SpaceNet 7 Baseline on AWSAdapted by Daniel Hogan from a post by Adam Van Etten and Nick Weir
InThe DownLinQbyDaniel Hogan·Aug 13, 2020The SpaceNet Change and Object Tracking (SCOT) MetricDaniel Hogan and Adam Van Etten
InThe DownLinQbyDaniel Hogan·Jul 16, 2020Introducing the Solaris Multimodal Preprocessing LibraryWithin the latest release of Solaris, available now, is a new resource for geospatial deep learning: the Solaris Multimodal Preprocessing…A response icon1A response icon1
InThe DownLinQbyDaniel Hogan·Mar 5, 2020The SpaceNet 6 BaselinePreface: SpaceNet LLC is a nonprofit organization dedicated to accelerating open source, artificial intelligence applied research for…A response icon2A response icon2
InThe DownLinQbyDaniel Hogan·Feb 13, 2020SAR 201: An Introduction to Synthetic Aperture Radar, Part 2By Daniel Hogan (In-Q-Tel CosmiQ Works) and Jason Brown (Capella Space) with the CosmiQ Works and Capella teams.
InThe DownLinQbyDaniel Hogan·Feb 5, 2020SAR 101: An Introduction to Synthetic Aperture RadarBy Daniel Hogan (In-Q-Tel CosmiQ Works) and Jason Brown (Capella Space) with the CosmiQ Works and Capella teams.A response icon4A response icon4
InThe DownLinQbyDaniel Hogan·Oct 16, 2019Robustness of Limited Training Data: Part 5Deep learning models for interpreting satellite imagery show increasing performance as the amount of training data is increased. We’ve…
InThe DownLinQbyDaniel Hogan·Sep 25, 2019Robustness of Limited Training Data: Part 4When it comes to the relationship between geospatial neural network performance and the amount of training data, do geographic differences…
InThe DownLinQbyDaniel Hogan·Sep 17, 2019Robustness of Limited Training Data: Part 3When training a deep neural network to identify building footprints in satellite imagery, having more training data never hurts. But how…