Bayesian Multi-Object Tracking Using Motion Context from Multiple Objects

TUD-Crossing PETS09-S2L2 ETH-Jelmoli ETH-Linthescher ETH-Crossing AVG-TownCentre ADL-Rundle-1 ADL-Rundle-3 KITTI-16 KITTI-19

Short name:

RMOT

Benchmark:

Description:

Hardware:

3.5 GHz, 1 Core

Detector:

Public

Processing:

Online

Last submitted:

December 09, 2014 (3 years ago)

Published:

December 09, 2014 at 03:08:21 CET

Submissions:

2

Open source:

No

Project page / code:

n/a

Reference:

J. Yoon, H. Yang, J. Lim, K. Yoon. Bayesian Multi-Object Tracking Using Motion Context from Multiple Objects. In IEEE Winter Conference on Applications of Computer Vision (WACV), 2015.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
18.669.62.25.3 % 53.3 % 12,47336,8356841,2823.5 GHz, 1 CorePublic
IDF1ID PrecisionID Recall
32.643.326.1

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing62.864.673.00.21330.8 % 15.4 % 343621419
PETS09-S2L237.232.367.72.6429.5 % 14.3 % 1,1264,743190320
ETH-Jelmoli40.449.671.30.64517.8 % 33.3 % 2631,2192959
ETH-Linthescher13.119.371.90.11971.5 % 81.2 % 1427,5892648
ETH-Crossing16.820.373.60.1263.8 % 76.9 % 11813109
AVG-TownCentre5.526.666.92.82260.9 % 59.7 % 1,2605,42474171
ADL-Rundle-1-1.333.269.79.63221.9 % 21.9 % 4,7904,54196192
ADL-Rundle-320.630.971.64.1449.1 % 22.7 % 2,5745,388112115
KITTI-1637.652.970.80.9170.0 % 17.6 % 1828582151
KITTI-1917.837.865.51.1624.8 % 32.3 % 1,1983,11779224
Venice-118.832.871.22.01711.8 % 35.3 % 8932,7813374

Raw data:


RMOT