Multi-object Tracking based on Union Model

MOT17-01-DPM MOT17-03-DPM MOT17-06-DPM MOT17-07-DPM MOT17-08-DPM MOT17-12-DPM MOT17-14-DPM MOT17-01-FRCNN MOT17-03-FRCNN MOT17-06-FRCNN

Short name:

yt_face

Benchmark:

Description:

n/a

Hardware:

Nvidia m40

Detector:

Public

Processing:

Online

Last submitted:

May 21, 2018 (6 months ago)

Published:

May 21, 2018 at 05:09:28 CET

Submissions:

1

Open source:

No

Project page / code:

n/a

Reference:

Anonymous submission

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
52.676.51.323.0 % 35.9 % 23,894241,4892,0472,827Nvidia m40Public
IDF1ID PrecisionID Recall
51.567.741.6

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
MOT17-01-DPM35.439.871.40.52416.7 % 41.7 % 2433,9062038
MOT17-03-DPM55.245.875.22.714823.6 % 16.2 % 4,05442,451343403
MOT17-06-DPM48.756.173.40.622224.8 % 45.9 % 7585,21470116
MOT17-07-DPM43.140.774.31.16010.0 % 40.0 % 5358,99187103
MOT17-08-DPM28.831.779.00.67613.2 % 47.4 % 37714,5997077
MOT17-12-DPM42.653.977.40.69120.9 % 46.2 % 5554,3873335
MOT17-14-DPM27.740.274.61.21649.1 % 53.0 % 93112,3745278
MOT17-01-FRCNN31.043.576.43.12429.2 % 33.3 % 1,4073,0251929
MOT17-03-FRCNN62.358.677.70.614828.4 % 15.5 % 94038,280191258
MOT17-06-FRCNN53.959.878.10.722228.4 % 33.8 % 8634,49773110
MOT17-07-FRCNN37.241.175.52.46011.7 % 28.3 % 1,2099,285122199
MOT17-08-FRCNN23.131.279.21.57611.8 % 51.3 % 91515,2825567
MOT17-12-FRCNN40.555.179.00.49115.4 % 51.6 % 3784,7641420
MOT17-14-FRCNN23.839.072.03.31648.5 % 45.7 % 2,46511,482131223
MOT17-01-SDP45.951.074.52.42437.5 % 20.8 % 1,0732,3843260
MOT17-03-SDP76.663.777.31.014852.0 % 9.5 % 1,54922,671297404
MOT17-06-SDP56.260.476.01.022239.6 % 32.9 % 1,2223,85287116
MOT17-07-SDP50.747.176.92.06026.7 % 25.0 % 9937,24499121
MOT17-08-SDP32.936.878.41.27618.4 % 44.7 % 73213,334107136
MOT17-12-SDP41.355.678.90.99120.9 % 45.1 % 8024,2632244
MOT17-14-SDP39.350.574.02.516411.0 % 33.5 % 1,8939,204123190

Raw data:

n/a


yt_face