Short name:
LMOT_Tracker
Detector:
Private
Description:
Reference:
R. Mostafa, H. Baraka, A. Bayoumi. LMOT: Efficient Light-Weight Detection and Tracking in Crowds. In IEEE Access, 2022.
Last submitted:
January 13, 2022 (3 years ago)
Published:
January 13, 2022 at 19:06:01 CET
Submissions:
1
Project page / code:
Open source:
No
Hardware:
RTX2070 Mobile GPU
Runtime:
29.6 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT16 | 73.2 | 72.3 | 56.8 | 334 (44.0) | 129 (17.0) | 12,798 | 35,454 | 80.6 | 92.0 | 55.8 | 58.2 | 61.5 | 73.4 | 64.4 | 73.5 | 80.8 | 2.2 | 669 (0.0) | 990 (0.0) |
Detailed performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT16-01 | 45.4 | 46.0 | 39.4 | 9 | 5 | 901 | 2,560 | 60.0 | 81.0 | 39.2 | 40.2 | 44.4 | 67.2 | 47.0 | 63.5 | 79.4 | 2.0 | 31 | 46 |
| MOT16-03 | 88.8 | 82.9 | 65.0 | 125 | 1 | 2,774 | 8,816 | 91.6 | 97.2 | 61.7 | 68.8 | 67.4 | 76.1 | 73.5 | 78.0 | 81.4 | 1.8 | 103 | 238 |
| MOT16-06 | 61.0 | 69.3 | 54.2 | 88 | 55 | 1,029 | 3,347 | 71.0 | 88.8 | 56.9 | 51.9 | 63.2 | 75.6 | 58.2 | 72.8 | 82.1 | 0.9 | 123 | 127 |
| MOT16-07 | 50.9 | 48.5 | 39.7 | 19 | 2 | 2,344 | 5,531 | 66.1 | 82.2 | 35.8 | 44.9 | 39.3 | 66.2 | 52.8 | 65.6 | 80.5 | 4.7 | 133 | 183 |
| MOT16-08 | 49.8 | 49.5 | 40.9 | 25 | 1 | 3,656 | 4,565 | 72.7 | 76.9 | 35.9 | 47.5 | 43.1 | 56.8 | 58.1 | 61.4 | 77.9 | 5.8 | 176 | 217 |
| MOT16-12 | 51.4 | 62.3 | 50.3 | 24 | 35 | 603 | 3,403 | 59.0 | 89.0 | 56.3 | 45.1 | 62.2 | 74.9 | 49.6 | 74.9 | 84.1 | 0.7 | 23 | 31 |
| MOT16-14 | 52.4 | 65.6 | 44.5 | 44 | 30 | 1,491 | 7,232 | 60.9 | 88.3 | 48.8 | 41.1 | 53.0 | 72.1 | 45.5 | 66.0 | 76.7 | 2.0 | 80 | 148 |
Raw data: