LMOT_Tracker: LMOT: Efficient Light-Weight Detection and Tracking in Crowds
MOT20-04
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 (2 years ago)
Published:
January 13, 2022 at 19:26:45 CET
Submissions:
1
Project page / code:
Open source:
No
Hardware:
RTX2070 Mobile GPU
Runtime:
22.4 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT20 | 59.1 | 61.1 | 47.5 | 312 (25.1) | 286 (23.0) | 13,526 | 196,673 | 62.0 | 96.0 | 46.8 | 48.4 | 51.5 | 73.7 | 51.2 | 79.2 | 83.1 | 3.0 | 1,398 (0.0) | 2,446 (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 |
| MOT20-04 | 56.8 | 61.3 | 46.0 | 74 | 156 | 1,109 | 116,835 | 57.4 | 99.3 | 45.9 | 46.1 | 48.9 | 78.6 | 47.7 | 82.5 | 83.8 | 0.5 | 386 | 565 |
| MOT20-06 | 62.9 | 61.8 | 50.6 | 124 | 73 | 5,498 | 43,278 | 67.4 | 94.2 | 49.5 | 51.9 | 55.8 | 70.2 | 55.7 | 77.9 | 82.9 | 5.5 | 541 | 1,062 |
| MOT20-07 | 70.7 | 61.6 | 50.4 | 66 | 7 | 3,479 | 6,042 | 81.7 | 88.6 | 43.1 | 59.2 | 50.2 | 63.4 | 67.5 | 73.2 | 82.3 | 5.9 | 169 | 192 |
| MOT20-08 | 55.8 | 59.0 | 46.0 | 48 | 50 | 3,440 | 30,518 | 60.6 | 93.2 | 46.6 | 45.5 | 52.5 | 69.7 | 49.0 | 75.4 | 81.8 | 4.3 | 302 | 627 |
Raw data:
