Benchmark:
MOT16 |
Short name:
RAR16pub
Detector:
Public
Description:
Recurrent autoregressive networks tracking framework with public detections
Reference:
K. Fang, Y. Xiang, X. Li, S. Savarese. Recurrent Autoregressive Networks for Online Multi-Object Tracking. In The IEEE Winter Conference on Applications of Computer Vision (WACV), 2018.
Last submitted:
March 24, 2017 (5 years ago)
Published:
July 23, 2018 at 09:33:19 CET
Submissions:
1
Project page / code:
Open source:
No
Hardware:
Titan X, 1.5GHZ, 1Core
Runtime:
0.9 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 | 45.9 | 48.8 | 36.5 | 100 (13.2) | 318 (41.9) | 6,871 | 91,173 | 50.0 | 93.0 | 36.3 | 36.8 | 38.1 | 75.8 | 38.9 | 72.4 | 78.5 | 1.2 | 648 (13.0) | 1,992 (39.8) |
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 | 26.3 | 33.6 | 25.9 | 2 | 13 | 41 | 4,663 | 27.1 | 97.7 | 33.7 | 19.9 | 35.1 | 75.6 | 20.4 | 73.5 | 77.5 | 0.1 | 12 | 49 |
| MOT16-03 | 54.6 | 52.8 | 39.7 | 34 | 22 | 3,492 | 43,679 | 58.2 | 94.6 | 36.8 | 43.0 | 38.6 | 76.0 | 45.4 | 73.7 | 78.6 | 2.3 | 301 | 1,072 |
| MOT16-06 | 45.2 | 53.9 | 38.9 | 31 | 107 | 414 | 5,821 | 49.5 | 93.2 | 42.6 | 35.7 | 44.8 | 74.8 | 37.6 | 70.8 | 77.0 | 0.3 | 86 | 205 |
| MOT16-07 | 39.9 | 45.9 | 32.8 | 6 | 19 | 659 | 9,074 | 44.4 | 91.7 | 33.7 | 32.2 | 35.3 | 73.5 | 34.0 | 70.2 | 77.5 | 1.3 | 73 | 188 |
| MOT16-08 | 29.8 | 35.1 | 28.1 | 6 | 24 | 971 | 10,688 | 36.1 | 86.2 | 29.0 | 27.2 | 30.5 | 76.7 | 29.2 | 69.6 | 80.1 | 1.6 | 89 | 166 |
| MOT16-12 | 39.0 | 48.3 | 37.0 | 12 | 40 | 634 | 4,380 | 47.2 | 86.1 | 40.1 | 34.2 | 42.1 | 75.8 | 37.3 | 68.0 | 78.6 | 0.7 | 43 | 131 |
| MOT16-14 | 26.6 | 39.1 | 27.4 | 9 | 93 | 660 | 12,868 | 30.4 | 89.5 | 33.3 | 22.5 | 34.6 | 76.1 | 23.6 | 69.5 | 77.8 | 0.9 | 44 | 181 |
Raw data: