Short name:
GLMBFairMOT
Detector:
Private
Description:
Visual multi-object tracking with re-identification and occlusion handling using labeled random finite sets
Reference:
L. Van Ma, T. Nguyen, C. Shim, D. Kim, N. Ha, M. Jeon. Visual multi-object tracking with re-identification and occlusion handling using labeled random finite sets. In Pattern Recognition, 2024.
Last submitted:
August 30, 2024 (2 months ago)
Published:
August 30, 2024 at 07:53:09 CET
Submissions:
1
Project page / code:
Open source:
No
Hardware:
CPU AMD Ryzen Threadripper 2950X 16-Core Processor
Runtime:
149.3 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT20 | 67.7 | 67.3 | 54.2 | 679 (54.7) | 180 (14.5) | 29,597 | 134,534 | 74.0 | 92.8 | 53.4 | 55.2 | 56.9 | 76.6 | 60.2 | 75.5 | 81.9 | 6.6 | 2,911 (0.0) | 5,486 (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 | 83.5 | 78.6 | 63.2 | 483 | 33 | 11,663 | 32,648 | 88.1 | 95.4 | 60.0 | 66.7 | 63.6 | 79.8 | 72.0 | 77.9 | 82.4 | 5.6 | 813 | 1,887 |
| MOT20-06 | 49.6 | 49.5 | 39.7 | 88 | 79 | 8,328 | 57,495 | 56.7 | 90.0 | 37.9 | 41.9 | 41.1 | 69.0 | 45.6 | 72.4 | 80.8 | 8.3 | 1,129 | 2,091 |
| MOT20-07 | 75.1 | 69.6 | 56.5 | 70 | 1 | 2,140 | 5,811 | 82.4 | 92.7 | 52.8 | 61.1 | 56.8 | 76.8 | 67.3 | 75.7 | 82.0 | 3.7 | 288 | 381 |
| MOT20-08 | 39.7 | 49.1 | 38.7 | 38 | 67 | 7,466 | 38,580 | 50.2 | 83.9 | 41.8 | 36.1 | 45.1 | 70.6 | 40.5 | 67.7 | 81.0 | 9.3 | 681 | 1,127 |
Raw data: