Benchmark:
MOT15 |
Short name:
RAR15pub
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 27, 2017 (7 years ago)
Published:
July 23, 2018 at 09:34:33 CET
Submissions:
2
Project page / code:
Open source:
No
Hardware:
Titan X, 1.5GHZ, 1Core
Runtime:
5.4 Hz
Benchmark performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| MOT15 | 35.1 | 45.4 | 33.2 | 94 (13.0) | 305 (42.3) | 6,771 | 32,717 | 46.7 | 80.9 | 34.9 | 31.9 | 37.8 | 69.9 | 35.5 | 61.4 | 75.1 | 1.2 | 381 (8.1) | 1,523 (32.6) |
Detailed performance:
| Sequence | MOTA | IDF1 | HOTA | MT | ML | FP | FN | Rcll | Prcn | AssA | DetA | AssRe | AssPr | DetRe | DetPr | LocA | FAF | ID Sw. | Frag |
| ADL-Rundle-1 | 26.3 | 47.4 | 34.8 | 8 | 11 | 2,135 | 4,690 | 49.6 | 68.4 | 40.2 | 30.3 | 43.1 | 72.9 | 37.5 | 51.7 | 76.0 | 4.3 | 31 | 168 |
| ADL-Rundle-3 | 39.6 | 43.1 | 31.3 | 4 | 10 | 729 | 5,344 | 47.4 | 86.9 | 29.4 | 33.7 | 30.7 | 72.4 | 36.2 | 66.4 | 75.9 | 1.2 | 64 | 148 |
| AVG-TownCentre | 30.3 | 51.3 | 37.6 | 42 | 96 | 917 | 4,038 | 43.5 | 77.2 | 47.9 | 29.7 | 53.3 | 67.6 | 33.2 | 58.9 | 73.8 | 2.0 | 25 | 237 |
| ETH-Crossing | 37.6 | 55.1 | 36.9 | 5 | 14 | 7 | 619 | 38.3 | 98.2 | 48.1 | 28.5 | 50.7 | 77.5 | 29.3 | 75.1 | 78.6 | 0.0 | 0 | 0 |
| ETH-Jelmoli | 40.6 | 58.3 | 39.9 | 9 | 13 | 486 | 1,004 | 60.4 | 75.9 | 41.7 | 38.4 | 47.7 | 67.1 | 46.3 | 58.1 | 76.6 | 1.1 | 18 | 79 |
| ETH-Linthescher | 26.9 | 38.2 | 28.9 | 12 | 132 | 199 | 6,307 | 29.4 | 92.9 | 38.4 | 21.9 | 41.0 | 75.6 | 22.6 | 71.5 | 77.6 | 0.2 | 23 | 64 |
| KITTI-16 | 41.2 | 57.7 | 38.4 | 0 | 3 | 174 | 809 | 52.4 | 83.7 | 41.0 | 36.1 | 45.1 | 63.5 | 39.9 | 63.7 | 75.3 | 0.8 | 18 | 66 |
| KITTI-19 | 31.3 | 50.1 | 33.5 | 3 | 14 | 819 | 2,809 | 47.4 | 75.6 | 34.8 | 32.8 | 39.1 | 61.0 | 36.3 | 57.8 | 71.2 | 0.8 | 41 | 300 |
| PETS09-S2L2 | 49.6 | 36.4 | 28.4 | 4 | 3 | 798 | 3,923 | 59.3 | 87.8 | 19.7 | 41.2 | 21.0 | 67.1 | 44.4 | 65.7 | 74.0 | 1.8 | 135 | 357 |
| TUD-Crossing | 70.1 | 77.8 | 52.6 | 6 | 1 | 46 | 273 | 75.2 | 94.7 | 54.0 | 51.2 | 57.1 | 73.3 | 55.5 | 69.9 | 77.0 | 0.2 | 11 | 29 |
| Venice-1 | 26.0 | 39.0 | 27.9 | 1 | 8 | 461 | 2,901 | 36.4 | 78.3 | 31.6 | 24.7 | 33.6 | 67.2 | 27.5 | 59.0 | 75.7 | 1.0 | 15 | 75 |
Raw data: