Planet partners with SpaceNet for multi-temporal urban development challenge

The main aim of the challenge is to develop better methods to track building construction overtime using Planet imagery mosaics

Planet partners with SpaceNet for multi-temporal urban development challenge
Planet partners with SpaceNet for multi-temporal urban development challenge

Planet partners with SpaceNet for multi-temporal urban development challenge

Planet partners with SpaceNet to support the SpaceNet7 Multi-Temporal Urban Development Challenge. The main aim of the challenge is to develop better methods to track building construction overtime using Planet imagery mosaics. Infrastructure development, disaster preparedness, epidemic prevention are a few efforts that can be supported with rapid and accurate remote-sensing of infrastructure.

SpaceNet is dedicated to accelerating the research and application of open source AI technology for geospatial applications. Free, precision-labeled, electro-optical and synthetic aperture radar satellite imagery datasets are offered by SpaceNet. They are running challenges with prizes to foster emerging analytical frameworks.

A dramatic expansion of the availability of open source data of building footprints and road networks for the geospatial machine learning community were led by the previous competitions.

The partnership between SpaceNet and Planet will allow the seventh competition to focus on discovery of change events directly. Access to a spatio-temporal dataset at a scale and cadence that was previously unavailable to the broader research community has unlocked this technical challenge. This competition will allow researchers to experiment with new change detection methods, not feasible on smaller datasets.

The Multi-Temporal Urban Development Challenge will commence on August 31 and will last two months. The results will be announced in December 2020 at the AI conference, Neural Information Processing Systems (NeurIPS).

Planet aims to image the entire world every day, make change visible, accessible and actionable.The key to realize this promise is to have the right algorithms and tools in order to change imagery into useful insights.