Submission details of TEAM_YW

Name TEAM_YW
Paper Link http://elvera.nue.tu-berlin.de/files/1517Bochinski2017.pdf
Code Link N/A
MOTA 0.205834
MOTP 0.254732
IDs 11466
False Positives 104613
False Negatives 643705
Input N/A
Runtime 0.001 s
Environment CPU
Abstract Tracking-by-detection is a common approach to multiobject tracking. With ever increasing performances of object detectors, the basis for a tracker becomes much more reliable. In combination with commonly higher frame rates, this poses a shift in the challenges for a successful tracker. That shift enables the deployment of much simpler tracking algorithms which can compete with more sophisticated approaches at a fraction of the computational cost. We present such an algorithm and show with thorough experiments its potential using a wide range of object detectors. The proposed method can easily run at 100K fps while outperforming the state-of-the-art on the DETRAC vehicle tracking dataset

Detailed results

Per-sequence results

sequence name MOTA MOTP IDs False Positives False Negatives
cubberly-auditorium-2019-04-22_1 24.12 26.45 193 2281 7566
discovery-walk-2019-02-28_0 20.15 26.75 146 1504 8985
discovery-walk-2019-02-28_1 28.59 25.14 230 1724 8391
food-trucks-2019-02-12_0 37.90 22.80 388 1427 37989
gates-ai-lab-2019-04-17_0 1.93 21.68 444 12595 12285
gates-basement-elevators-2019-01-17_0 49.00 22.48 158 757 5411
gates-foyer-2019-01-17_0 21.38 22.32 117 5018 2875
gates-to-clark-2019-02-28_0 30.30 23.63 49 503 1226
hewlett-class-2019-01-23_0 54.63 21.13 250 1158 6094
hewlett-class-2019-01-23_1 82.03 20.55 69 267 587
huang-2-2019-01-25_1 24.18 24.80 77 1598 3352
huang-intersection-2019-01-22_0 9.48 27.49 484 8270 33691
indoor-coupa-cafe-2019-02-06_0 18.30 26.41 809 6547 46239
lomita-serra-intersection-2019-01-30_0 13.34 26.70 157 2906 14783
meyer-green-2019-03-16_1 7.59 26.90 267 3091 18274
nvidia-aud-2019-01-25_0 13.13 28.86 473 6613 21691
nvidia-aud-2019-04-18_1 35.15 20.71 45 1280 3853
nvidia-aud-2019-04-18_2 10.23 28.81 107 4349 4684
outdoor-coupa-cafe-2019-02-06_0 7.09 28.59 427 5480 40694
quarry-road-2019-02-28_0 -13.08 26.89 107 2780 3794
serra-street-2019-01-30_0 1.40 28.29 527 6328 35560
stlc-111-2019-04-19_1 55.58 21.03 74 521 2178
stlc-111-2019-04-19_2 47.30 21.53 45 640 1228
tressider-2019-03-16_2 13.18 25.94 222 3112 19807
tressider-2019-04-26_0 15.86 27.58 1869 7513 91360
tressider-2019-04-26_1 27.93 24.62 1421 5675 131665
tressider-2019-04-26_3 20.76 28.02 2311 10676 79443
total 20.58 25.47 11466 104613 643705