Introduction:
The SAT-MTB-MLSC dataset consists of 2,232 satellite videos for training and 1,317 satellite videos for testing with overall 141,960 frames. There are 14 classes of static ground things (airport, sea, buildings, runway, parking lot, railway, overpass, coastline,... Introduction:
The SAT-MTB-MLSC dataset consists of 2,232 satellite videos for training and 1,317 satellite videos for testing with overall 141,960 frames. There are 14 classes of static ground things (airport, sea, buildings, runway, parking lot, railway, overpass, coastl...2023-04-10
Introduction:
SatSOT consists of 105 satellite video sequences: 26 trains, 65 cars, 9 planes and 5 ships with 27,664 frames in total.
The average video length of SatSOT is 263 frames. The shortest video consists of 120 frames, while the longest one contains 750 frames...2021-06-07
SAT-MTB:A Multi-task Benchmark Dataset for Satellite Videos: Object Detection, Tracking, and Segmentation
Introduction:SatMTB contains 249 videos with more than 50,000 sequential frames. The object categories are divided into two levels of coarse and fine granularity, with aircraft, ships, vehicles, trains and their corresponding 14 types of fine-grained objects annotated frame by...2022-11-21