Benchmark:
MOT16 |
Short name:
FWT
Detector:
Public
Description:
n/a
Reference:
R. Henschel, L. Leal-Taixé, D. Cremers, B. Rosenhahn. Fusion of Head and Full-Body Detectors for Multi-Object Tracking. In Trajnet CVPRW, 2018.
Last submitted:
June 15, 2017 (7 years ago)
Published:
March 17, 2017 at 22:54:36 CET
Submissions:
3
Project page / code:
n/a
Open source:
No
Hardware:
3GHZ, 8 Cores
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 | 47.8 | 44.3 | 35.7 | 145 (19.1) | 290 (38.2) | 8,886 | 85,487 | 53.1 | 91.6 | 32.7 | 39.1 | 35.3 | 70.7 | 41.8 | 72.0 | 78.7 | 1.5 | 852 (16.0) | 1,534 (28.9) |
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 | 33.6 | 28.2 | 24.6 | 5 | 10 | 460 | 3,741 | 41.5 | 85.2 | 21.2 | 28.8 | 25.5 | 59.5 | 31.3 | 64.4 | 76.7 | 1.0 | 46 | 52 |
MOT16-03 | 55.7 | 47.3 | 38.0 | 41 | 22 | 4,574 | 41,326 | 60.5 | 93.3 | 32.6 | 44.6 | 35.0 | 71.4 | 47.5 | 73.3 | 78.8 | 3.0 | 403 | 898 |
MOT16-06 | 51.8 | 55.4 | 42.7 | 46 | 98 | 358 | 5,110 | 55.7 | 94.7 | 45.4 | 40.4 | 48.6 | 75.4 | 42.7 | 72.7 | 78.3 | 0.3 | 90 | 121 |
MOT16-07 | 40.3 | 40.0 | 31.0 | 7 | 19 | 823 | 8,844 | 45.8 | 90.1 | 29.0 | 33.6 | 31.1 | 70.2 | 35.7 | 70.1 | 77.7 | 1.6 | 76 | 157 |
MOT16-08 | 35.1 | 28.2 | 24.3 | 9 | 19 | 985 | 9,739 | 41.8 | 87.7 | 18.7 | 31.9 | 21.8 | 61.5 | 34.2 | 71.9 | 80.8 | 1.6 | 136 | 137 |
MOT16-12 | 44.6 | 56.5 | 44.0 | 23 | 36 | 781 | 3,788 | 54.3 | 85.2 | 49.6 | 39.2 | 53.3 | 73.5 | 43.6 | 68.4 | 79.9 | 0.9 | 26 | 41 |
MOT16-14 | 24.7 | 35.1 | 27.2 | 14 | 86 | 905 | 12,939 | 30.0 | 86.0 | 33.8 | 22.1 | 36.2 | 71.2 | 23.4 | 67.1 | 77.4 | 1.2 | 75 | 128 |
Raw data: