Leveraging single for multi-target tracking using a novel trajectory overlap affinity measure
Short name:
TO
Benchmark:
Description:
n/a
Hardware:
1.6 GHZ, 8 Cores
Detector:
Public
Processing:
Batch
Last submitted:
September 02, 2015 (4 years ago)
Published:
April 19, 2017 at 14:44:27 CET
Submissions:
1
Open source:
No
Project page / code:
n/a
Reference:
S. Manen, R. Timofte, D. Dai, L. Gool. Leveraging single for multi-target tracking using a novel trajectory overlap affinity measure. In 2016 IEEE Winter Conference on Applications of Computer Vision (WACV), 2016.
Benchmark performance:
| MOTA | MOTP | FAF | MT | ML | FP | FN | ID Sw. | Frag | Specifications | Detector |
| 25.7 | 72.2 | 0.8 | 4.3 % | 57.4 % | 4,779 | 40,511 | 383 | 600 | 1.6 GHZ, 8 Cores | Public |
| IDF1 | ID Precision | ID Recall |
| 32.7 | 55.5 | 23.2 |
Detailed performance:
| Sequence | MOTA | IDF1 | MOTP | FAF | GT | MT | ML | FP | FN | ID Sw | Frag |
| TUD-Crossing | 57.2 | 51.5 | 73.1 | 0.1 | 13 | 23.1 % | 23.1 % | 26 | 429 | 17 | 17 |
| PETS09-S2L2 | 40.7 | 30.7 | 70.6 | 1.2 | 42 | 7.1 % | 19.0 % | 522 | 5,049 | 145 | 168 |
| ETH-Jelmoli | 32.8 | 43.0 | 75.1 | 0.3 | 45 | 13.3 % | 42.2 % | 139 | 1,559 | 7 | 21 |
| ETH-Linthescher | 8.3 | 13.3 | 77.3 | 0.0 | 197 | 0.5 % | 87.3 % | 24 | 8,162 | 7 | 8 |
| ETH-Crossing | 16.7 | 22.6 | 76.5 | 0.0 | 26 | 0.0 % | 73.1 % | 2 | 830 | 4 | 4 |
| AVG-TownCentre | 18.8 | 25.9 | 70.0 | 0.5 | 226 | 2.7 % | 61.5 % | 209 | 5,533 | 63 | 131 |
| ADL-Rundle-1 | 17.9 | 38.4 | 73.4 | 4.7 | 32 | 18.8 % | 31.3 % | 2,336 | 5,270 | 33 | 49 |
| ADL-Rundle-3 | 31.6 | 36.2 | 74.1 | 1.1 | 44 | 4.5 % | 29.5 % | 683 | 6,216 | 51 | 50 |
| KITTI-16 | 39.9 | 48.0 | 74.4 | 0.3 | 17 | 0.0 % | 11.8 % | 64 | 951 | 7 | 20 |
| KITTI-19 | 32.2 | 43.8 | 65.7 | 0.4 | 62 | 6.5 % | 33.9 % | 451 | 3,137 | 32 | 113 |
| Venice-1 | 18.6 | 25.2 | 74.4 | 0.7 | 17 | 0.0 % | 47.1 % | 323 | 3,375 | 17 | 19 |
Raw data:
