ALExTRAC: Affinity Learning by Exploring Temporal Reinforcement within Association Chains


Benchmark:

Short name:

ALExTRAC

Detector:

Public

Description:

Appearance only tracking through the use of a self-supervised affinity model.

Project page / code:

Reference:

A. Bewley, L. Ott, F. Ramos, B. Upcroft. ALExTRAC: Affinity Learning by Exploring Temporal Reinforcement within Association Chains. In International Conference on Robotics and Automation (ICRA), (to appear) 2016.

Processing:

Batch

Last submitted:

September 12, 2015 (4 years ago)

Published:

February 16, 2016 at 04:03:23 CET

Submissions:

1

Open source:

No

Hardware:

2.5 GHz, 4 cores

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN ID Sw.
2D MOT 201517.017.371.228.0378.09,23339,9331,859

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN ID Sw.
ADL-Rundle-15.113.571.35.07.03,5035,010321
ADL-Rundle-318.116.471.83.010.02,4205,523385
AVG-TownCentre13.321.370.03.0138.04425,637118
ETH-Crossing19.623.174.72.017.01378310
ETH-Jelmoli32.933.073.26.013.02831,33486
ETH-Linthescher15.417.874.13.0150.0947,36990
KITTI-1613.912.872.10.04.01051,27980
KITTI-1913.521.066.54.019.01,1283,266227
PETS09-S2L227.512.970.60.011.04496,153385
TUD-Crossing51.241.073.02.02.03945940
Venice-112.512.271.90.07.07573,120117

Raw data:


ALExTRAC