MOT17

MOT17 Challenge. All MOT16 sequences are used with a new, more accurate ground truth. Each sequences is provided with 3 sets of detections: DPM, Faster-RCNN, and SDP.

Jump to download

Training Set

SampleName FPS Resolution Length Tracks Boxes DensityDescriptionSourceRef.
1MOT17-13-SDP251920x1080750 (00:30)1101164215.5Filmed from a bus on a busy intersectionlink[1]
2MOT17-11-SDP301920x1080900 (00:30)75943610.5Forward moving camera in a busy shopping malllink[1]
3MOT17-10-SDP301920x1080654 (00:22)571283919.6A pedestrian scene filmed at night by a moving cameralink[2]
4MOT17-09-SDP301920x1080525 (00:18)26532510.1A pedestrian street scene filmed from a low angle.link[2]
5MOT17-05-SDP14640x480837 (01:00)13369178.3Street scene from a moving platformlink[3]
6MOT17-04-SDP301920x10801050 (00:35)834755745.3Pedestrian street at night, elevated viewpointlink[1]
7MOT17-02-SDP301920x1080600 (00:20)621858131.0People walking around a large square.link[2]
8MOT17-13-FRCNN251920x1080750 (00:30)1101164215.5Filmed from a bus on a busy intersectionlink[1]
9MOT17-11-FRCNN301920x1080900 (00:30)75943610.5Forward moving camera in a busy shopping malllink[1]
10MOT17-10-FRCNN301920x1080654 (00:22)571283919.6A pedestrian scene filmed at night by a moving cameralink[2]
11MOT17-09-FRCNN301920x1080525 (00:18)26532510.1A pedestrian street scene filmed from a low angle.link[2]
12MOT17-05-FRCNN14640x480837 (01:00)13369178.3Street scene from a moving platformlink[3]
13MOT17-04-FRCNN301920x10801050 (00:35)834755745.3Pedestrian street at night, elevated viewpointlink[1]
14MOT17-02-FRCNN301920x1080600 (00:20)621858131.0People walking around a large square.link[2]
15MOT17-13-DPM251920x1080750 (00:30)1101164215.5Filmed from a bus on a busy intersectionlink[1]
16MOT17-11-DPM301920x1080900 (00:30)75943610.5Forward moving camera in a busy shopping malllink[1]
17MOT17-10-DPM301920x1080654 (00:22)571283919.6A pedestrian scene filmed at night by a moving cameralink[2]
18MOT17-09-DPM301920x1080525 (00:18)26532510.1A pedestrian street scene filmed from a low angle.link[2]
19MOT17-05-DPM14640x480837 (01:00)13369178.3Street scene from a moving platformlink[3]
20MOT17-04-DPM301920x10801050 (00:35)834755745.3Pedestrian street at night, elevated viewpointlink[1]
21MOT17-02-DPM301920x1080600 (00:20)621858131.0People walking around a large square.link[2]
Total 15948 frm.
(645 s.)
1638 336891 21.1

Test Set

SampleName FPS Resolution Length Tracks Boxes DensityDescriptionSourceRef.
1MOT17-14-SDP251920x1080750 (00:30)1641848324.6Filmed from a bus on a busy intersectionlink[1]
2MOT17-12-SDP301920x1080900 (00:30)9186679.6Forward moving camera in a busy shopping malllink[1]
3MOT17-08-SDP301920x1080625 (00:21)762112433.8A crowded pedestrian street, stationary cameralink[2]
4MOT17-07-SDP301920x1080500 (00:17)601689333.8A busy pedestrian street filmed at eye level by a moving cameralink[2]
5MOT17-06-SDP14640x4801194 (01:25)222117849.9Street scene from a moving platformlink[3]
6MOT17-03-SDP301920x10801500 (00:50)14810467569.8Pedestrian street at night, elevated viewpointlink[1]
7MOT17-01-SDP301920x1080450 (00:15)24645014.3People walking around a large square.link[2]
8MOT17-14-FRCNN251920x1080750 (00:30)1641848324.6Filmed from a bus on a busy intersectionlink[1]
9MOT17-12-FRCNN301920x1080900 (00:30)9186679.6Forward moving camera in a busy shopping malllink[1]
10MOT17-08-FRCNN301920x1080625 (00:21)762112433.8A crowded pedestrian street, stationary cameralink[2]
11MOT17-07-FRCNN301920x1080500 (00:17)601689333.8A busy pedestrian street filmed at eye level by a moving cameralink[2]
12MOT17-06-FRCNN14640x4801194 (01:25)222117849.9Street scene from a moving platformlink[3]
13MOT17-03-FRCNN301920x10801500 (00:50)14810467569.8Pedestrian street at night, elevated viewpointlink[1]
14MOT17-01-FRCNN301920x1080450 (00:15)24645014.3People walking around a large square.link[2]
15MOT17-14-DPM251920x1080750 (00:30)1641848324.6Filmed from a bus on a busy intersectionlink[1]
16MOT17-12-DPM301920x1080900 (00:30)9186679.6Forward moving camera in a busy shopping malllink[1]
17MOT17-08-DPM301920x1080625 (00:21)762112433.8A crowded pedestrian street, stationary cameralink[2]
18MOT17-07-DPM301920x1080500 (00:17)601689333.8A busy pedestrian street filmed at eye level by a moving cameralink[2]
19MOT17-06-DPM14640x4801194 (01:25)222117849.9Street scene from a moving platformlink[3]
20MOT17-03-DPM301920x10801500 (00:50)14810467569.8Pedestrian street at night, elevated viewpointlink[1]
21MOT17-01-DPM301920x1080450 (00:15)24645014.3People walking around a large square.link[2]
Total 17757 frm.
(744 s.)
2355 564228 31.8


Download

Get all data (5.5 GB)
Get detections and labels only (9.7 MB)
Get development kit (0.5 MB)

Note that the data contains the same set of sequences (frames) as MOT16 three times. For convenience you may download the entire data which will extract in correct folder structure. Alternatively, you may re-use the MOT16 sequences (frames) locally. Important: Both the ground truth and the detection set is new for MOT17!

References:


[1] Milan, A., Leal-Taixé, L., Reid, I., Roth, S. & Schindler, K. MOT16: A Benchmark for Multi-Object Tracking. arXiv:1603.00831 [cs], 2016., (arXiv: 1603.00831).
[2] Leal-Taixé, L., Milan, A., Reid, I., Roth, S. & Schindler, K. MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking. arXiv:1504.01942 [cs], 2015., (arXiv: 1504.01942).
[3] Ess, A., Leibe, B. & Gool, L.V. Depth and Appearance for Mobile Scene Analysis. In Proceedings of the Eleventh IEEE International Conference on Computer Vision, 2007.