Multi Object Tracking using Convolutional Neural Networks for Person ReIdentification

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

Short name:

ReID_CNN

Benchmark:

Description:

n/a

Hardware:

Nvidia GTX 1080Ti and i7 7th Gen

Detector:

Private

Processing:

Batch

Last submitted:

March 30, 2019 (7 months ago)

Published:

March 30, 2019 at 11:57:06 CET

Submissions:

2

Open source:

No

Project page / code:

n/a

Reference:

Anonymous submission

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
38.974.31.322.6 % 33.1 % 7,47329,4046811,158Nvidia GTX 1080Ti and i7 7th GenPrivate
IDF1ID PrecisionID Recall
41.553.034.1

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing76.356.576.20.31369.2 % 0.0 % 541881923
PETS09-S2L249.029.972.51.44211.9 % 7.1 % 6104,121186319
ETH-Jelmoli25.552.875.62.64533.3 % 26.7 % 1,1257362861
ETH-Linthescher55.857.277.11.119736.5 % 33.5 % 1,3372,50799123
ETH-Crossing70.954.579.40.42638.5 % 26.9 % 782001421
AVG-TownCentre32.045.469.71.822612.8 % 38.1 % 8213,95192192
ADL-Rundle-135.540.473.52.03221.9 % 31.3 % 1,0074,94456104
ADL-Rundle-337.242.077.52.24427.3 % 15.9 % 1,3724,93579101
KITTI-1616.128.268.00.4170.0 % 52.9 % 881,3231640
KITTI-1918.627.467.70.3621.6 % 54.8 % 3543,9554298
Venice-129.426.572.61.41717.6 % 29.4 % 6272,5445076

Raw data:

n/a


ReID_CNN