MOT20Det

Pedestrian Detection Challenge. This benchmark contains 8 challenging video sequences (4 train, 4 test) in unconstrained environments. Tracking and evaluation are done in image coordinates. All sequences have been annotated with high accuracy, strictly following a well-defined protocol.

Training Set

Sample Name FPS Resolution Length Tracks BoxesDensityDescriptionSourceRef.
MOT20-05251654x10803315 (02:13)1211751330226.6Crowded square by night time.link[1]
MOT20-03251173x8802405 (01:36)735356728148.3People leaving entrance of stadium by night time, elevated viewpoint.link[1]
MOT20-02251920x10802782 (01:51)29620221572.7Crowded indoor train station.link[1]
MOT20-01251920x1080429 (00:17)902664762.1Crowded indoor train station.link[1]
Total 8931 frm.
(357 s.)
2332 1336920 149.7

Test Set

Sample Name FPS Resolution Length Tracks BoxesDensityDescriptionSourceRef.
MOT20-08251920x734806 (00:32)279145301180.3A pedestrian street scene.link[1]
MOT20-07251920x1080585 (00:23)1264109670.2Crowded indoor train station.link[1]
MOT20-06251920x7341008 (00:40)368207543205.9A pedestrian street scene.link[1]
MOT20-04251545x10802080 (01:23)728371525178.6Crowded square by night time.link[1]
Total 4479 frm.
(178 s.)
1501 765465 170.9


Download

Get all data (4.7 GB)
Get files (no img) only (11.5 MB)
Development Kit

References:


[1] Dendorfer, P., Rezatofighi, H., Milan, A., Shi, J., Cremers, D., Reid, I., Roth, S., Schindler, K. & Leal-Taixé, L. MOT20: A benchmark for multi object tracking in crowded scenes. arXiv:2003.09003[cs], 2020., (arXiv: 2003.09003).