Spacenet Challenge Round 2 - Buildings (CosmiQ Works, Radiant Solutions, NVIDIA, May 2017)Ħ85k building footprints, 3/8band Worldview-3 imagery (0.3m res.), 5 cities, SpaceNet Challenge Asset Library Open the atom feed downloadlinks with Firefox etc., not Chrome.ĭenmark LPIS agricultural field boundaries (Denmark Department for Agriculture)Ģ93 crop/vegetation catgeories, 600k parcels, yearly dataset for 2008-2018ĬrowdAI Mapping Challenge (Humanity & Inclusion NGO, May 2018)īuildings footprints, RGB satellite imagery, COCO data format Netherlands LPIS agricultural field boundaries (Netherlands Department for Economic Affairs)Ģ94 crop/vegetation catgeories, 780k parcels, yearly dataset for 2009-2018. Open AI Challenge: Tanzania (WeRobotics & Wordlbank, Nov 2018)īuilding footprints & 3 building conditions, RGB UAV imagery - Link to data 2019Īirbus Ship Detection Challenge (Airbus, Nov 2018)ġ31k ships, 104k train / 88k test image chips, satellite imagery (1.5m res.), raster mask labels in in run-length encoding format, Kaggle kernels. Bi-cubicly resampled to same number of pixels in each image to counter courser native resolution with higher off-nadir angles, Paper: Weir et al. Spacenet Challenge Round 4 - Off-nadir (CosmiQ Works, DigitalGlobe, Radiant Solutions, AWS, Dec 2018)ġ26k building footprints (Atlanta), 27 WorldView 2 images (0.3m res.) from 7-54 degrees off-nadir angle.
Microsoft BuildingFootprints Canada & USA & Uganda/Tanzania (Microsoft, Mar 2019)ġ2.6mil (Canada) & 125.2mil (USA) & 17.9mil (Uganda/Tanzania) building footprints, GeoJSON format, delineation based on Bing imagery using ResNet34 architecture. ĥ50k building footprints & 4 damage scale categories, 20 global locations and 7 disaster types (wildfire, landslides, dam collapses, volcanic eruptions, earthquakes/tsunamis, wind, flooding), Worldview-3 imagery (0.3m res.), pre-trained baseline model. XView 2 Building Damage Asessment Challenge (DIUx, Nov 2019). Spacenet Challenge Round 6 - Multi-Sensor All Weather Mapping (CosmiQ Works, Capella Space, Maxar, AWS, Intel, Feb 2020)Ĥ8k building footprints (enhanced 3DBAG dataset, building height attributes), Capella Space SAR data (0.5m res., four polarizations) & Worldview-3 imagery (0.3m res.), Rotterdam, Netherlands. 2020Īgriculture-Vision Database & CVPR 2020 challenge (UIUC,Īgricultural Pattern Analysis, 21k aerial farmland images (RGB-NIR, USA, 2019 season, 512x512px chips), label masks for 6 field anomaly patterns (Cloud shadow, Double plant, Planter skip, Standing Water, Waterway and Weed cluster). Synthetic (630k planes, 50k images) and real (14.7k planes, 253 Worldview-3 images (0.3m res.), 122 locations, 22 countries) plane annotations & properties and satellite images. RarePlanes: Synthetic Data Takes Flight (CosmiQ Works, A.I.Reverie, June 2020) res) timeseries for 2 years, 100 locations around the globe, for building footprint evolution & address propagation.
Monthly building footprints and Planet imagery (4m.
SpaceNet 7: Multi-Temporal Urban Development Challenge (CosmiQ Works, Planet, Aug 2020) Recent additions and ongoing competitions Newest datasets at the top of each category (Instance segmentation, object detection, semantic segmentation, scene classification, other). List of aerial and satellite imagery datasets with annotations for computer vision and deep learning. List of satellite image training datasets with annotations for computer vision and deep learningĬhrieke/awesome-satellite-imagery-datasets
0 kommentar(er)
