Submission details of Tracktor++_yw

Name Tracktor++_yw
Paper Link N/A
Code Link N/A
MOTA 0.317604
MOTP 0.226798
IDs 6514
False Positives 44142
False Negatives 602197
Input N/A
Runtime 0.5 s
Environment 1 GPU (Titan X)
Abstract The problem of tracking multiple objects in a video se- quence poses several challenging tasks. For tracking-by- detection, these include object re-identification, motion pre- diction and dealing with occlusions. We present a tracker (without bells and whistles) that accomplishes tracking without specifically targeting any of these tasks, in partic- ular, we perform no training or optimization on tracking data. To this end, we exploit the bounding box regression of an object detector to predict the position of an object in the next frame, thereby converting a detector into a Tracktor. We demonstrate the potential of Tracktor and provide a new state-of-the-art on three multi-object tracking benchmarks by extending it with a straightforward re-identification and camera motion compensation. We then perform an analysis on the performance and failure cases of several state-of-the-art tracking methods in comparison to our Tracktor. Surprisingly, none of the dedicated tracking methods are considerably better in deal- ing with complex tracking scenarios, namely, small and occluded objects or missing detections. However, our ap- proach tackles most of the easy tracking scenarios. There- fore, we motivate our approach as a new tracking paradigm and point out promising future research directions. Over- all, Tracktor yields superior tracking performance than any current tracking method and our analysis exposes remain- ing and unsolved tracking challenges to inspire future re- search directions.

Detailed results

Per-sequence results

sequence name MOTA MOTP IDs False Positives False Negatives
cubberly-auditorium-2019-04-22_1 46.22 23.88 74 524 6517
discovery-walk-2019-02-28_0 34.58 23.04 105 543 8064
discovery-walk-2019-02-28_1 38.81 22.75 97 574 8193
food-trucks-2019-02-12_0 47.11 20.48 315 658 32927
gates-ai-lab-2019-04-17_0 43.09 18.09 165 2606 11925
gates-basement-elevators-2019-01-17_0 56.82 21.03 83 230 5043
gates-foyer-2019-01-17_0 62.37 18.97 51 1213 2570
gates-to-clark-2019-02-28_0 53.27 18.28 23 256 913
hewlett-class-2019-01-23_0 59.56 18.95 172 413 6102
hewlett-class-2019-01-23_1 84.25 17.07 31 135 643
huang-2-2019-01-25_1 38.08 20.75 30 582 3493
huang-intersection-2019-01-22_0 22.53 25.84 268 3040 33020
indoor-coupa-cafe-2019-02-06_0 32.40 21.74 286 2190 41870
lomita-serra-intersection-2019-01-30_0 22.67 23.32 105 1604 14216
meyer-green-2019-03-16_1 22.03 25.13 142 1355 16755
nvidia-aud-2019-01-25_0 36.19 24.56 241 2206 18691
nvidia-aud-2019-04-18_1 46.96 17.72 31 528 3676
nvidia-aud-2019-04-18_2 52.49 24.98 47 628 4162
outdoor-coupa-cafe-2019-02-06_0 19.77 24.77 316 3941 35987
quarry-road-2019-02-28_0 21.04 27.27 39 983 3643
serra-street-2019-01-30_0 15.10 25.02 543 2864 33114
stlc-111-2019-04-19_1 66.15 18.03 39 378 1696
stlc-111-2019-04-19_2 62.62 18.63 36 341 980
tressider-2019-03-16_2 21.83 25.35 119 716 20000
tressider-2019-04-26_0 24.38 24.17 1024 4313 85203
tressider-2019-04-26_1 29.71 22.01 1072 4107 130148
tressider-2019-04-26_3 30.63 25.40 1060 7214 72646
total 31.76 22.68 6514 44142 602197