AFN: Tracklet Association Tracker:
MOT16-01
Benchmark:
MOT16 |
Short name:
AFN
Detector:
Public
Description:
n/a
Reference:
H. Shen, L. Huang, C. Huang, W. Xu. Tracklet Association Tracker: An End-to-End Learning-based Association Approach for Multi-Object Tracking. In CoRR, 2018.
Last submitted:
January 24, 2018 (6 years ago)
Published:
September 04, 2018 at 18:03:09 CET
Submissions:
1
Project page / code:
n/a
Open source:
No
Hardware:
2GHZ Xeon, Titan X GPU
Runtime:
0.6 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 | 49.0 | 48.2 | 38.6 | 145 (19.1) | 271 (35.7) | 9,508 | 82,506 | 54.7 | 91.3 | 36.5 | 41.1 | 39.9 | 74.2 | 44.2 | 73.7 | 80.8 | 1.6 | 899 (16.4) | 1,383 (25.3) |
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 | 29.8 | 39.0 | 30.5 | 5 | 9 | 754 | 3,689 | 42.3 | 78.2 | 33.1 | 28.5 | 37.0 | 61.9 | 32.2 | 59.5 | 76.3 | 1.7 | 46 | 87 |
| MOT16-03 | 57.0 | 49.5 | 40.4 | 43 | 23 | 4,349 | 40,132 | 61.6 | 93.7 | 35.2 | 46.8 | 37.9 | 75.6 | 50.0 | 76.0 | 81.3 | 2.9 | 469 | 624 |
| MOT16-06 | 51.9 | 59.1 | 44.7 | 49 | 95 | 445 | 5,043 | 56.3 | 93.6 | 47.7 | 42.0 | 57.3 | 68.5 | 44.8 | 74.4 | 80.5 | 0.4 | 63 | 166 |
| MOT16-07 | 46.8 | 49.7 | 38.1 | 10 | 14 | 735 | 7,856 | 51.9 | 92.0 | 37.7 | 38.8 | 40.0 | 75.8 | 41.4 | 73.4 | 80.2 | 1.5 | 89 | 147 |
| MOT16-08 | 32.4 | 37.2 | 31.1 | 10 | 24 | 973 | 10,226 | 38.9 | 87.0 | 32.4 | 30.0 | 36.7 | 69.7 | 32.2 | 72.0 | 81.5 | 1.6 | 109 | 135 |
| MOT16-12 | 39.5 | 53.3 | 42.0 | 16 | 35 | 950 | 4,036 | 51.3 | 81.8 | 48.8 | 36.3 | 54.4 | 70.9 | 41.3 | 65.7 | 80.5 | 1.1 | 33 | 52 |
| MOT16-14 | 30.1 | 41.8 | 30.9 | 12 | 71 | 1,302 | 11,524 | 37.7 | 84.2 | 34.7 | 27.7 | 37.0 | 74.0 | 30.0 | 67.2 | 79.1 | 1.7 | 90 | 172 |
Raw data:
