Benchmark:
MOT17 |
Short name:
MASS
Detector:
Public
Description:
n/a
Reference:
H. Karunasekera, H. Wang, H. Zhang. Multiple Object Tracking With Attention to Appearance, Structure, Motion and Size. In IEEE Access, 2019.
Last submitted:
April 16, 2019 (5 years ago)
Published:
April 04, 2019 at 12:28:45 CET
Submissions:
3
Project page / code:
n/a
Open source:
No
Hardware:
Intel Xeon(R) CPU E5-1630 v3 @ 3.70GHz × 8 (1 Core used)
Runtime:
17.1 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 | 46.9 | 46.0 | 36.8 | 399 (16.9) | 856 (36.3) | 25,733 | 269,116 | 52.3 | 92.0 | 35.4 | 38.5 | 38.8 | 70.2 | 41.2 | 72.4 | 79.4 | 1.4 | 4,478 (85.6) | 11,994 (229.3) |
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 | 25.2 | 30.4 | 24.3 | 2 | 12 | 107 | 4,664 | 27.7 | 94.3 | 29.5 | 20.1 | 31.5 | 69.6 | 20.7 | 70.5 | 76.6 | 0.2 | 55 | 169 |
| MOT17-01-FRCNN | 24.9 | 35.0 | 32.3 | 6 | 7 | 1,589 | 3,215 | 50.2 | 67.1 | 32.7 | 32.3 | 35.3 | 72.8 | 40.6 | 54.3 | 78.8 | 3.5 | 39 | 78 |
| MOT17-01-SDP | 35.8 | 35.5 | 30.3 | 7 | 4 | 1,348 | 2,675 | 58.5 | 73.7 | 25.1 | 36.9 | 28.3 | 66.8 | 45.2 | 56.9 | 77.3 | 3.0 | 118 | 251 |
| MOT17-03-DPM | 45.1 | 41.1 | 31.3 | 20 | 35 | 3,769 | 53,056 | 49.3 | 93.2 | 27.4 | 35.9 | 29.7 | 68.2 | 38.0 | 71.8 | 78.3 | 2.5 | 602 | 2,270 |
| MOT17-03-FRCNN | 58.0 | 51.4 | 40.9 | 38 | 25 | 1,988 | 41,658 | 60.2 | 96.9 | 36.9 | 45.9 | 39.6 | 72.8 | 48.1 | 77.5 | 80.7 | 1.3 | 347 | 869 |
| MOT17-03-SDP | 73.6 | 59.2 | 47.8 | 70 | 13 | 1,492 | 25,519 | 75.6 | 98.2 | 40.7 | 56.4 | 44.5 | 70.2 | 59.4 | 77.1 | 80.3 | 1.0 | 636 | 1,893 |
| MOT17-06-DPM | 43.0 | 44.5 | 32.7 | 21 | 110 | 183 | 6,433 | 45.4 | 96.7 | 32.9 | 32.7 | 41.6 | 60.3 | 33.9 | 72.2 | 77.0 | 0.2 | 106 | 376 |
| MOT17-06-FRCNN | 54.6 | 57.4 | 44.8 | 55 | 56 | 593 | 4,599 | 61.0 | 92.4 | 43.9 | 46.0 | 50.9 | 68.0 | 49.3 | 74.7 | 80.6 | 0.5 | 161 | 339 |
| MOT17-06-SDP | 55.9 | 61.3 | 45.0 | 66 | 66 | 744 | 4,298 | 63.5 | 91.0 | 45.2 | 45.0 | 53.0 | 65.4 | 49.1 | 70.4 | 78.8 | 0.6 | 157 | 340 |
| MOT17-07-DPM | 32.9 | 35.8 | 26.6 | 4 | 30 | 365 | 10,841 | 35.8 | 94.3 | 27.1 | 26.3 | 28.8 | 70.0 | 27.3 | 71.8 | 77.2 | 0.7 | 128 | 638 |
| MOT17-07-FRCNN | 32.5 | 38.1 | 30.9 | 7 | 15 | 1,691 | 9,461 | 44.0 | 81.5 | 31.4 | 31.6 | 33.8 | 67.3 | 34.9 | 64.6 | 77.4 | 3.4 | 254 | 566 |
| MOT17-07-SDP | 45.1 | 43.4 | 34.8 | 12 | 15 | 1,127 | 7,934 | 53.0 | 88.8 | 32.4 | 37.7 | 34.7 | 71.0 | 41.1 | 68.8 | 78.4 | 2.3 | 209 | 668 |
| MOT17-08-DPM | 24.2 | 27.5 | 23.7 | 6 | 42 | 270 | 15,625 | 26.0 | 95.3 | 27.5 | 20.5 | 30.0 | 71.1 | 21.0 | 76.9 | 81.2 | 0.4 | 123 | 275 |
| MOT17-08-FRCNN | 22.1 | 28.7 | 26.6 | 7 | 38 | 1,000 | 15,341 | 27.4 | 85.3 | 33.3 | 21.4 | 35.3 | 77.4 | 22.7 | 70.7 | 81.9 | 1.6 | 108 | 185 |
| MOT17-08-SDP | 28.6 | 30.8 | 27.0 | 10 | 35 | 779 | 14,044 | 33.5 | 90.1 | 28.5 | 25.9 | 31.9 | 70.9 | 27.2 | 73.0 | 80.5 | 1.2 | 261 | 465 |
| MOT17-12-DPM | 38.5 | 49.9 | 38.1 | 9 | 47 | 337 | 4,948 | 42.9 | 91.7 | 45.8 | 31.7 | 48.8 | 74.1 | 33.6 | 71.7 | 79.3 | 0.4 | 41 | 139 |
| MOT17-12-FRCNN | 36.0 | 49.4 | 38.9 | 14 | 39 | 820 | 4,681 | 46.0 | 82.9 | 45.1 | 33.6 | 51.4 | 72.6 | 37.4 | 67.5 | 80.7 | 0.9 | 43 | 122 |
| MOT17-12-SDP | 40.3 | 54.8 | 43.3 | 19 | 37 | 932 | 4,175 | 51.8 | 82.8 | 50.4 | 37.4 | 53.7 | 76.5 | 42.2 | 67.4 | 80.9 | 1.0 | 69 | 176 |
| MOT17-14-DPM | 19.1 | 27.1 | 20.3 | 5 | 106 | 632 | 14,192 | 23.2 | 87.2 | 24.4 | 17.0 | 26.3 | 66.1 | 17.7 | 66.6 | 76.6 | 0.8 | 133 | 460 |
| MOT17-14-FRCNN | 15.7 | 32.9 | 26.0 | 11 | 70 | 3,425 | 11,656 | 36.9 | 66.6 | 27.3 | 25.5 | 30.5 | 59.9 | 29.6 | 53.4 | 74.1 | 4.6 | 500 | 833 |
| MOT17-14-SDP | 29.5 | 40.1 | 29.2 | 10 | 54 | 2,542 | 10,101 | 45.3 | 76.7 | 28.6 | 30.2 | 31.4 | 63.5 | 34.5 | 58.4 | 75.5 | 3.4 | 388 | 882 |
Raw data: