CRF_TRA_17:
MOT17-01-DPM
Benchmark:
MOT17 |
Short name:
CRF_TRA_17
Detector:
Public
Description:
n/a
Reference:
Jun xiang, Chao Ma, Guohan Xu, Jianhua Hou, End-to-End Learning Deep CRF models for Multi-Object Tracking. In IEEE Transactions on Circuits and Systems for Video Technology, 2020
Last submitted:
June 20, 2019 (1 year ago)
Published:
January 17, 2019 at 15:12:54 CET
Submissions:
3
Project page / code:
n/a
Open source:
Yes
Hardware:
Runtime:
1.4 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
| MOT17 | 53.1 | 53.7 | 76.1 | 571 (24.2) | 724 (30.7) | 27,194 | 234,991 | 58.4 | 92.4 | 1.5 | 2,518 (43.2) | 4,918 (84.3) |
Detailed performance:
| Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
| MOT17-01-DPM | 47.6 | 63.0 | 73.1 | 8 | 7 | 484 | 2,889 | 55.2 | 88.0 | 1.1 | 9 | 44 |
| MOT17-01-FRCNN | 32.3 | 47.5 | 76.6 | 7 | 7 | 1,166 | 3,182 | 50.7 | 73.7 | 2.6 | 20 | 39 |
| MOT17-01-SDP | 43.7 | 51.7 | 73.6 | 10 | 4 | 1,399 | 2,189 | 66.1 | 75.3 | 3.1 | 42 | 90 |
| MOT17-03-DPM | 54.7 | 55.1 | 75.6 | 32 | 27 | 4,138 | 42,949 | 59.0 | 93.7 | 2.8 | 284 | 562 |
| MOT17-03-FRCNN | 60.7 | 59.1 | 77.1 | 44 | 22 | 3,267 | 37,709 | 64.0 | 95.3 | 2.2 | 209 | 380 |
| MOT17-03-SDP | 76.3 | 59.6 | 77.4 | 83 | 14 | 2,572 | 21,867 | 79.1 | 97.0 | 1.7 | 395 | 722 |
| MOT17-06-DPM | 46.5 | 59.4 | 75.2 | 40 | 101 | 511 | 5,732 | 51.4 | 92.2 | 0.4 | 57 | 158 |
| MOT17-06-FRCNN | 55.4 | 61.2 | 77.3 | 81 | 42 | 1,535 | 3,629 | 69.2 | 84.2 | 1.3 | 88 | 216 |
| MOT17-06-SDP | 59.0 | 63.8 | 75.9 | 104 | 59 | 1,295 | 3,433 | 70.9 | 86.6 | 1.1 | 108 | 177 |
| MOT17-07-DPM | 45.2 | 49.7 | 75.3 | 9 | 17 | 840 | 8,330 | 50.7 | 91.1 | 1.7 | 82 | 193 |
| MOT17-07-FRCNN | 39.6 | 41.7 | 73.1 | 6 | 12 | 994 | 9,051 | 46.4 | 88.8 | 2.0 | 154 | 310 |
| MOT17-07-SDP | 52.1 | 49.6 | 74.3 | 17 | 11 | 1,039 | 6,912 | 59.1 | 90.6 | 2.1 | 140 | 265 |
| MOT17-08-DPM | 31.7 | 43.3 | 79.3 | 15 | 34 | 978 | 13,374 | 36.7 | 88.8 | 1.6 | 67 | 84 |
| MOT17-08-FRCNN | 26.0 | 31.6 | 77.6 | 8 | 38 | 385 | 15,159 | 28.2 | 93.9 | 0.6 | 81 | 113 |
| MOT17-08-SDP | 30.0 | 32.4 | 75.5 | 9 | 34 | 760 | 13,856 | 34.4 | 90.5 | 1.2 | 167 | 233 |
| MOT17-12-DPM | 45.7 | 56.6 | 78.7 | 18 | 39 | 346 | 4,329 | 50.1 | 92.6 | 0.4 | 35 | 54 |
| MOT17-12-FRCNN | 41.4 | 55.1 | 76.6 | 13 | 42 | 479 | 4,572 | 47.2 | 89.5 | 0.5 | 24 | 51 |
| MOT17-12-SDP | 44.1 | 57.6 | 75.7 | 18 | 31 | 890 | 3,913 | 54.9 | 84.2 | 1.0 | 45 | 74 |
| MOT17-14-DPM | 31.2 | 43.3 | 74.8 | 14 | 73 | 909 | 11,721 | 36.6 | 88.2 | 1.2 | 91 | 158 |
| MOT17-14-FRCNN | 28.9 | 39.6 | 69.3 | 11 | 62 | 1,869 | 11,033 | 40.3 | 79.9 | 2.5 | 231 | 527 |
| MOT17-14-SDP | 42.2 | 49.0 | 69.8 | 24 | 48 | 1,338 | 9,162 | 50.4 | 87.4 | 1.8 | 189 | 468 |
Raw data:
