Benchmark:
MOT17 |
Short name:
PHD_DCM
Detector:
Public
Description:
n/a
Reference:
Z. Fu, F. Angelini, S. Naqvi, J. Chambers. GM-PHD Filter Based Online Multiple Human Tracking Using Deep Discriminative Correlation Matching. In 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018.
Last submitted:
October 07, 2017 (7 years ago)
Published:
November 19, 2020 at 12:25:12 CET
Submissions:
1
Project page / code:
n/a
Open source:
No
Hardware:
i7 3.5GHz, 4 Cores; GTX 1060
Runtime:
1.6 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 | 46.5 | 47.6 | 38.2 | 397 (16.9) | 875 (37.2) | 23,859 | 272,430 | 51.7 | 92.4 | 37.9 | 38.8 | 41.1 | 73.0 | 41.3 | 73.8 | 80.2 | 1.3 | 5,649 (109.2) | 9,298 (179.8) |
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 | 25.5 | 34.1 | 25.9 | 1 | 12 | 191 | 4,556 | 29.4 | 90.8 | 31.5 | 21.3 | 32.9 | 71.4 | 22.2 | 68.8 | 77.1 | 0.4 | 58 | 135 |
MOT17-01-FRCNN | 26.3 | 40.5 | 34.8 | 6 | 8 | 1,468 | 3,259 | 49.5 | 68.5 | 37.0 | 33.0 | 42.7 | 66.5 | 40.7 | 56.3 | 79.1 | 3.3 | 27 | 75 |
MOT17-01-SDP | 38.6 | 43.0 | 36.4 | 7 | 5 | 1,075 | 2,729 | 57.7 | 77.6 | 35.3 | 38.1 | 37.6 | 75.8 | 45.1 | 60.7 | 78.3 | 2.4 | 155 | 191 |
MOT17-03-DPM | 43.4 | 39.1 | 30.6 | 19 | 38 | 3,980 | 54,657 | 47.8 | 92.6 | 26.5 | 35.7 | 28.5 | 67.4 | 37.7 | 73.1 | 79.6 | 2.7 | 573 | 1,793 |
MOT17-03-FRCNN | 57.4 | 54.9 | 42.5 | 35 | 25 | 1,756 | 42,496 | 59.4 | 97.3 | 40.1 | 45.5 | 43.5 | 72.4 | 47.6 | 77.9 | 80.9 | 1.2 | 322 | 777 |
MOT17-03-SDP | 73.3 | 61.0 | 50.0 | 70 | 16 | 1,253 | 26,048 | 75.1 | 98.4 | 43.7 | 57.5 | 46.5 | 75.9 | 60.4 | 79.1 | 81.5 | 0.8 | 677 | 1,422 |
MOT17-06-DPM | 43.0 | 54.2 | 39.4 | 26 | 105 | 239 | 6,290 | 46.6 | 95.8 | 46.1 | 34.1 | 50.8 | 72.8 | 35.5 | 73.1 | 77.9 | 0.2 | 193 | 281 |
MOT17-06-FRCNN | 53.1 | 54.4 | 43.2 | 50 | 57 | 684 | 4,584 | 61.1 | 91.3 | 41.3 | 45.5 | 53.8 | 60.8 | 49.1 | 73.4 | 80.1 | 0.6 | 253 | 279 |
MOT17-06-SDP | 52.8 | 58.8 | 45.4 | 70 | 66 | 797 | 4,414 | 62.5 | 90.2 | 46.2 | 44.9 | 50.4 | 75.0 | 49.2 | 71.0 | 79.7 | 0.7 | 351 | 273 |
MOT17-07-DPM | 32.0 | 41.2 | 30.5 | 4 | 29 | 451 | 10,831 | 35.9 | 93.1 | 35.0 | 26.8 | 36.5 | 77.1 | 27.9 | 72.3 | 78.3 | 0.9 | 212 | 447 |
MOT17-07-FRCNN | 32.9 | 40.0 | 32.2 | 6 | 16 | 1,566 | 9,526 | 43.6 | 82.5 | 33.8 | 31.7 | 36.2 | 70.5 | 34.8 | 65.8 | 77.8 | 3.1 | 250 | 432 |
MOT17-07-SDP | 46.1 | 46.9 | 37.2 | 14 | 12 | 1,194 | 7,616 | 54.9 | 88.6 | 35.3 | 39.6 | 37.5 | 75.6 | 43.4 | 70.0 | 79.6 | 2.4 | 296 | 439 |
MOT17-08-DPM | 22.7 | 28.6 | 26.3 | 6 | 41 | 453 | 15,736 | 25.5 | 92.2 | 33.8 | 20.5 | 35.7 | 79.5 | 21.2 | 76.6 | 82.5 | 0.7 | 146 | 194 |
MOT17-08-FRCNN | 21.9 | 28.9 | 27.4 | 8 | 40 | 871 | 15,523 | 26.5 | 86.5 | 35.9 | 21.0 | 38.2 | 80.8 | 22.1 | 72.2 | 82.2 | 1.4 | 102 | 140 |
MOT17-08-SDP | 29.2 | 32.1 | 28.7 | 10 | 35 | 772 | 13,860 | 34.4 | 90.4 | 30.9 | 26.9 | 33.7 | 75.0 | 28.3 | 74.4 | 81.5 | 1.2 | 329 | 381 |
MOT17-12-DPM | 37.6 | 49.3 | 38.3 | 10 | 53 | 233 | 5,107 | 41.1 | 93.9 | 47.0 | 31.2 | 51.6 | 74.0 | 32.7 | 74.7 | 80.6 | 0.3 | 64 | 104 |
MOT17-12-FRCNN | 39.2 | 51.3 | 39.8 | 12 | 48 | 296 | 4,926 | 43.2 | 92.7 | 48.1 | 33.1 | 53.4 | 75.7 | 35.0 | 75.1 | 81.2 | 0.3 | 49 | 80 |
MOT17-12-SDP | 41.4 | 55.8 | 44.1 | 19 | 37 | 719 | 4,255 | 50.9 | 86.0 | 51.9 | 37.6 | 54.8 | 78.0 | 41.6 | 70.2 | 81.6 | 0.8 | 106 | 127 |
MOT17-14-DPM | 17.7 | 29.6 | 21.5 | 3 | 102 | 922 | 13,921 | 24.7 | 83.2 | 25.4 | 18.3 | 26.8 | 74.2 | 19.4 | 65.4 | 77.9 | 1.2 | 363 | 418 |
MOT17-14-FRCNN | 17.3 | 35.9 | 28.2 | 10 | 73 | 3,096 | 11,704 | 36.7 | 68.6 | 31.9 | 25.6 | 35.5 | 62.9 | 29.3 | 54.9 | 74.3 | 4.1 | 489 | 647 |
MOT17-14-SDP | 30.4 | 42.7 | 30.9 | 11 | 57 | 1,843 | 10,392 | 43.8 | 81.4 | 32.0 | 30.2 | 34.6 | 71.5 | 33.5 | 62.4 | 76.6 | 2.5 | 634 | 663 |
Raw data: