The new xView dataset released March 2018, is one the largest and most diverse public overhead object detection datasets available from around the world, annotated using bounding boxes. It contains more than 1 million labeled objects covering over 1400km² of the earth’s surface. Objects belong to 60 classes, including fine-grained classes. xView targets four computer vision frontiers: improving low-resolution and multi-scale recognition, improving learning efficiency, pushing the limit of discoverable object categories, and improving detection of fine-grained classes.

xView Challenge is an initiative of the US Department of Defense’s Defense Innovation Unit (DIU) and the US National Geospatial-Intelligence Agency (NGA), encouraging scientists and mathematicians to create algorithms to automatically identify objects from satellite images.

Company Type: Enterprise

Region: US & Canada

Industry Category: Government Satellite Imagery

Data set: xView Dataset

ML Platform Library: TensorFlow Models for Object Detection

Research: xView: Objects in Context in Overhead Imagery