Benchmark:
MOT17 |
Short name:
SORT17
Detector:
Public
Description:
SORT is a barebones implementation of a visual multiple object tracking framework based on the Hungarian data association using IOU as an affinity score and Kalman filter state estimation. Using the open-source code on MOT17 without re-tuning any parameters.
Reference:
A. Bewley, Z. Ge, L. Ott, F. Ramos, B. Upcroft. Simple online and realtime tracking. In 2016 IEEE International Conference on Image Processing (ICIP), 2016.
Last submitted:
July 21, 2019 (5 years ago)
Published:
July 21, 2019 at 19:43:20 CET
Submissions:
1
Project page / code:
Open source:
Yes
Hardware:
2.5 GHz, 4 cores
Runtime:
143.3 Hz
Benchmark performance:
Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
MOT17 | 43.1 | 39.8 | 34.0 | 295 (12.5) | 997 (42.3) | 28,398 | 287,582 | 49.0 | 90.7 | 31.8 | 37.0 | 33.1 | 79.6 | 39.5 | 73.1 | 80.5 | 1.6 | 4,852 (99.0) | 7,127 (145.4) |
Detailed performance:
Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
MOT17-01-DPM | 22.0 | 23.7 | 19.9 | 2 | 11 | 434 | 4,501 | 30.2 | 81.8 | 18.7 | 21.5 | 19.4 | 73.8 | 23.1 | 62.6 | 76.4 | 1.0 | 93 | 128 |
MOT17-01-FRCNN | 28.9 | 35.5 | 33.3 | 6 | 8 | 1,174 | 3,374 | 47.7 | 72.4 | 34.0 | 32.9 | 35.9 | 78.4 | 39.1 | 59.4 | 79.5 | 2.6 | 35 | 52 |
MOT17-01-SDP | 44.3 | 40.1 | 34.7 | 6 | 6 | 413 | 3,101 | 51.9 | 89.0 | 32.6 | 37.3 | 33.9 | 79.0 | 40.6 | 69.6 | 79.1 | 0.9 | 81 | 130 |
MOT17-03-DPM | 35.7 | 28.6 | 25.8 | 21 | 35 | 13,505 | 52,560 | 49.8 | 79.4 | 19.0 | 35.5 | 19.5 | 79.9 | 40.3 | 64.3 | 78.8 | 9.0 | 1,229 | 1,677 |
MOT17-03-FRCNN | 55.7 | 46.5 | 38.4 | 33 | 28 | 1,265 | 44,645 | 57.3 | 97.9 | 33.6 | 44.4 | 35.1 | 79.8 | 46.2 | 78.9 | 81.4 | 0.8 | 425 | 612 |
MOT17-03-SDP | 69.7 | 53.5 | 45.7 | 59 | 16 | 732 | 30,346 | 71.0 | 99.0 | 38.2 | 55.1 | 39.9 | 79.5 | 57.4 | 80.1 | 81.9 | 0.5 | 617 | 1,212 |
MOT17-06-DPM | 35.9 | 37.9 | 29.6 | 12 | 123 | 432 | 6,885 | 41.6 | 91.9 | 29.8 | 29.9 | 31.0 | 77.5 | 31.6 | 69.8 | 77.9 | 0.4 | 232 | 259 |
MOT17-06-FRCNN | 47.9 | 51.3 | 40.1 | 28 | 83 | 218 | 5,766 | 51.1 | 96.5 | 40.6 | 39.7 | 42.6 | 80.0 | 41.3 | 78.1 | 81.2 | 0.2 | 157 | 209 |
MOT17-06-SDP | 50.2 | 52.3 | 40.1 | 37 | 83 | 246 | 5,472 | 53.6 | 96.2 | 40.3 | 40.2 | 42.5 | 78.3 | 42.1 | 75.6 | 80.0 | 0.2 | 154 | 179 |
MOT17-07-DPM | 25.9 | 24.9 | 20.4 | 4 | 26 | 1,577 | 10,684 | 36.8 | 79.7 | 16.0 | 26.7 | 16.4 | 77.6 | 29.2 | 63.4 | 77.1 | 3.2 | 261 | 342 |
MOT17-07-FRCNN | 32.6 | 34.3 | 28.6 | 4 | 22 | 849 | 10,324 | 38.9 | 88.6 | 28.9 | 29.2 | 30.4 | 75.3 | 31.0 | 70.6 | 78.7 | 1.7 | 213 | 322 |
MOT17-07-SDP | 42.4 | 40.5 | 33.9 | 9 | 22 | 459 | 9,099 | 46.1 | 94.4 | 33.0 | 35.1 | 34.5 | 78.6 | 36.9 | 75.5 | 80.7 | 0.9 | 178 | 282 |
MOT17-08-DPM | 17.4 | 25.0 | 23.2 | 6 | 40 | 1,605 | 15,684 | 25.8 | 77.2 | 26.4 | 20.6 | 27.5 | 80.3 | 22.2 | 66.5 | 81.1 | 2.6 | 163 | 179 |
MOT17-08-FRCNN | 21.9 | 28.4 | 26.5 | 6 | 42 | 566 | 15,847 | 25.0 | 90.3 | 35.2 | 20.1 | 36.3 | 84.0 | 20.9 | 75.6 | 83.1 | 0.9 | 84 | 115 |
MOT17-08-SDP | 27.2 | 30.1 | 27.8 | 10 | 38 | 215 | 15,028 | 28.9 | 96.6 | 33.3 | 23.4 | 34.5 | 83.5 | 24.0 | 80.4 | 83.3 | 0.3 | 131 | 181 |
MOT17-12-DPM | 31.8 | 41.0 | 32.9 | 10 | 45 | 890 | 4,949 | 42.9 | 80.7 | 35.2 | 30.8 | 36.9 | 77.9 | 34.5 | 64.8 | 80.0 | 1.0 | 70 | 98 |
MOT17-12-FRCNN | 37.2 | 50.0 | 38.9 | 11 | 49 | 284 | 5,122 | 40.9 | 92.6 | 47.9 | 31.8 | 50.7 | 80.3 | 33.4 | 75.6 | 81.8 | 0.3 | 38 | 55 |
MOT17-12-SDP | 42.6 | 55.0 | 43.7 | 15 | 45 | 229 | 4,702 | 45.7 | 94.5 | 53.1 | 36.0 | 55.3 | 82.0 | 37.8 | 78.0 | 82.9 | 0.3 | 46 | 66 |
MOT17-14-DPM | 16.4 | 24.1 | 19.2 | 5 | 104 | 922 | 14,400 | 22.1 | 81.6 | 22.4 | 16.7 | 23.0 | 79.5 | 17.6 | 64.9 | 77.9 | 1.2 | 139 | 236 |
MOT17-14-FRCNN | 19.5 | 30.5 | 23.6 | 7 | 90 | 1,613 | 12,975 | 29.8 | 77.3 | 26.3 | 22.0 | 27.8 | 71.6 | 23.7 | 61.6 | 75.4 | 2.2 | 290 | 429 |
MOT17-14-SDP | 29.1 | 35.5 | 25.7 | 4 | 81 | 770 | 12,118 | 34.4 | 89.2 | 26.7 | 25.1 | 27.6 | 78.7 | 26.6 | 68.9 | 78.3 | 1.0 | 216 | 364 |
Raw data: