TUD-Crossing
Benchmark:
Short name:
DCO_X
Detector:
Public
Description:
We formulate multi-target tracking as a discrete-continuous optimization problem that handles both data association and trajectory estimation in its respective natural domain and allows leveraging powerful methods for multi-model fitting. Data association is performed using discrete optimization with label costs, yielding near optimality. Trajectory estimation is posed as a continuous fitting problem which is used in turn to update the label costs.
Reference:
A. Milan, K. Schindler, S. Roth. Multi-Target Tracking by Discrete-Continuous Energy Minimization. In IEEE PAMI, 2016.
Last submitted:
June 02, 2015 (5 years ago)
Published:
July 28, 2015 at 17:01:02 CET
Submissions:
2
Project page / code:
Open source:
Yes
Hardware:
2.7 GHz, 1 Core
Runtime:
0.3 Hz
Benchmark performance:
Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
2D MOT 2015 | 19.6 | 31.5 | 71.4 | 37 (5.1) | 396 (54.9) | 10,652 | 38,232 | 37.8 | 68.5 | 1.8 | 521 (13.8) | 819 (21.7) |
Detailed performance:
Sequence | MOTA | IDF1 | MOTP | MT | ML | FP | FN | Recall | Precision | FAF | ID Sw. | Frag |
ADL-Rundle-1 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
ADL-Rundle-3 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
AVG-TownCentre | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
ETH-Crossing | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
ETH-Jelmoli | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
ETH-Linthescher | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
KITTI-16 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
KITTI-19 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
PETS09-S2L2 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
TUD-Crossing | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
Venice-1 | 0.0 | 0.0 | 0.0 | 0 | 0 | 0 | 0 | 0.0 | 0.0 | 0.0 | 0 | 0 |
Raw data: