EarthVision will feature the 6th SpaceNet (https://spacenet.ai/) challenge.  
 
 
 

SpaceNet 6: Multi-Sensor All-Weather Mapping
Spanning three years while featuring five unique datasets and challenges, SpaceNet has continued to focus on different aspects of applying machine learning to solve difficult foundational mapping problems. The SpaceNet team has open sourced ~27,000 sq. km of imagery, 811,000 building footprints, and ~20,000 km of road labels across 10 cities. Using these datasets, the SpaceNet public challenges have focused on several different geospatial aspects including: high resolution building footprint mapping, the effects of look angle on model performance, and creating routable road networks. As we look ahead to our next challenge, SpaceNet 6, we will push into a new frontier and an underexplored data modality: Synthetic Aperture Radar (SAR).

 
A New Modality – The SpaceNet 6 Dataset
Synthetic Aperture Radar sensors are unique as they can penetrate clouds and can collect during all weather conditions. Furthermore, radar satellites do not require illumination and can capture data during both the day and the night. Consequently, overhead collects from SAR satellites could be particularly valuable in the quest to aid disaster response, when weather and cloud cover often obstruct traditional electro-optical sensors.

Despite these many advantages, there is little open data available to researchers to explore the effectiveness of SAR for such applications, particularly at ultra-high resolutions. Thanks to new SpaceNet partner Capella Space, SpaceNet will be the first to feature open-source half-meter SAR data in a challenge setting. Additionally, to compliment to the SAR collect the dataset will also feature half-meter electro-optical imagery from Maxar’s WorldView satellites. Our area of interest for this challenge will be centered over the largest port in Europe: Rotterdam, the Netherlands. The collected area features thousands of buildings, vehicles, and boats of various sizes, which will make for an effective test bed for SAR and the fusion of these two types of data.

 
Competition Structure and CVPR EarthVision Workshop
In the SpaceNet 6 challenge, participants will be asked to automatically extract building footprints with computer vision and artificial intelligence (AI) algorithms using a combination of these two diverse remote sensing datasets. For training data, participants will be allowed to leverage both the electro-optical and SAR datasets. However, for testing models and scoring performance only a subset of the data will be made available. We hope that such a structure will incentivize new data fusion methods and other approaches such as domain adaptation.

 
The SpaceNet 6 challenge will run for approximately two months, with an anticipated launch date of February 2020. Data preprocessing is ongoing now and will be publicly released as the competition launch nears. Evaluation will again rely on the SpaceNet metric, implementing the F1 score, which represents the harmonic average of precision and recall. Finally, following the challenge SpaceNet’s best participants will be invited to share their work at one of CVPR’s most respected workshops: EarthVision 2020. The SpaceNet 6 challenge will run for approximately two months, running March 16 – May 1, 2020. The dataset may be found here.
References:
[1] Dukai, B. (Balázs) (2018) 3D Registration of Buildings and Addresses (BAG). 4TU.Centre for Research Data. Dataset. https://doi.org/10.4121/uuid:f1f9759d-024a-492a-b821-07014dd6131c