AFN17: Tracklet Association Tracker:
MOT17-14-FRCNN
Benchmark:
MOT17 |
Short name:
AFN17
Detector:
Public
Description:
n/a
Reference:
H. Shen, L. Huang, C. Huang, W. Xu. Tracklet Association Tracker: An End-to-End Learning-based Association Approach for Multi-Object Tracking. In CoRR, 2018.
Last submitted:
March 31, 2018 (2 years ago)
Published:
April 10, 2018 at 14:06:21 CET
Submissions:
2
Project page / code:
n/a
Open source:
No
Hardware:
NVIDIA TITAN X
Runtime:
1.8 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
| MOT17 | 51.5 | 46.9 | 77.6 | 485 (20.6) | 836 (35.5) | 22,391 | 248,420 | 56.0 | 93.4 | 1.3 | 2,593 (46.3) | 4,308 (77.0) |
Detailed performance:
| Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
| MOT17-01-DPM | 35.7 | 45.2 | 75.7 | 4 | 10 | 220 | 3,899 | 39.6 | 92.1 | 0.5 | 31 | 59 |
| MOT17-01-FRCNN | 31.5 | 37.1 | 75.9 | 8 | 9 | 1,256 | 3,130 | 51.5 | 72.6 | 2.8 | 33 | 51 |
| MOT17-01-SDP | 36.6 | 41.0 | 75.7 | 9 | 3 | 1,668 | 2,369 | 63.3 | 71.0 | 3.7 | 52 | 108 |
| MOT17-03-DPM | 56.6 | 51.7 | 77.5 | 32 | 28 | 2,025 | 43,086 | 58.8 | 96.8 | 1.4 | 320 | 516 |
| MOT17-03-FRCNN | 59.7 | 47.5 | 77.5 | 38 | 25 | 2,027 | 39,848 | 61.9 | 97.0 | 1.4 | 328 | 430 |
| MOT17-03-SDP | 73.9 | 49.9 | 78.6 | 77 | 14 | 2,733 | 24,101 | 77.0 | 96.7 | 1.8 | 438 | 775 |
| MOT17-06-DPM | 44.5 | 49.0 | 77.0 | 45 | 107 | 888 | 5,568 | 52.7 | 87.5 | 0.7 | 80 | 136 |
| MOT17-06-FRCNN | 54.4 | 54.5 | 77.2 | 52 | 77 | 380 | 4,889 | 58.5 | 94.8 | 0.3 | 108 | 158 |
| MOT17-06-SDP | 57.9 | 60.1 | 77.5 | 80 | 62 | 753 | 4,102 | 65.2 | 91.1 | 0.6 | 105 | 176 |
| MOT17-07-DPM | 43.0 | 46.3 | 76.6 | 7 | 23 | 410 | 9,119 | 46.0 | 95.0 | 0.8 | 92 | 150 |
| MOT17-07-FRCNN | 38.4 | 40.0 | 75.9 | 7 | 15 | 830 | 9,448 | 44.1 | 90.0 | 1.7 | 127 | 220 |
| MOT17-07-SDP | 48.8 | 45.1 | 76.8 | 16 | 13 | 1,143 | 7,375 | 56.3 | 89.3 | 2.3 | 126 | 219 |
| MOT17-08-DPM | 25.0 | 29.9 | 79.5 | 7 | 37 | 657 | 15,093 | 28.6 | 90.2 | 1.1 | 91 | 116 |
| MOT17-08-FRCNN | 24.3 | 30.5 | 80.0 | 8 | 37 | 617 | 15,301 | 27.6 | 90.4 | 1.0 | 66 | 89 |
| MOT17-08-SDP | 29.7 | 33.5 | 79.4 | 10 | 35 | 614 | 14,100 | 33.3 | 92.0 | 1.0 | 139 | 193 |
| MOT17-12-DPM | 40.7 | 52.5 | 77.3 | 14 | 43 | 423 | 4,691 | 45.9 | 90.4 | 0.5 | 28 | 40 |
| MOT17-12-FRCNN | 40.9 | 53.5 | 78.5 | 13 | 46 | 440 | 4,659 | 46.2 | 90.1 | 0.5 | 22 | 37 |
| MOT17-12-SDP | 42.3 | 54.5 | 78.1 | 20 | 38 | 857 | 4,109 | 52.6 | 84.2 | 1.0 | 38 | 53 |
| MOT17-14-DPM | 30.3 | 40.8 | 76.6 | 10 | 84 | 610 | 12,211 | 33.9 | 91.1 | 0.8 | 56 | 112 |
| MOT17-14-FRCNN | 26.1 | 34.9 | 73.5 | 11 | 78 | 1,441 | 12,041 | 34.9 | 81.7 | 1.9 | 182 | 330 |
| MOT17-14-SDP | 36.1 | 48.8 | 74.4 | 17 | 52 | 2,399 | 9,281 | 49.8 | 79.3 | 3.2 | 131 | 340 |
Raw data:
