Multi-Object Tracking with Quadruplet Convolutional Neural Networks

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

Short name:

QuadMOT

Benchmark:

Description:

n/a

Hardware:

3 GHZ, 1 Core, TITAN X

Detector:

Public

Processing:

Batch

Last submitted:

November 14, 2016 (1 year ago)

Published:

November 23, 2016 at 04:26:12 CET

Submissions:

2

Open source:

No

Project page / code:

n/a

Reference:

J. Son, M. Baek, M. Cho, B. Han. Multi-Object Tracking with Quadruplet Convolutional Neural Networks. In CVPR, 2017.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
33.873.41.412.9 % 36.9 % 7,89832,0617031,4303 GHZ, 1 Core, TITAN XPublic
IDF1ID PrecisionID Recall
40.453.532.5

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing72.159.874.50.01346.2 % 0.0 % 52881521
PETS09-S2L249.028.872.61.64216.7 % 7.1 % 6863,947285380
ETH-Jelmoli42.355.475.60.94517.8 % 28.9 % 3791,0651950
ETH-Linthescher28.740.375.80.21977.1 % 62.9 % 2706,0563881
ETH-Crossing33.453.077.70.2267.7 % 57.7 % 4162705
AVG-TownCentre30.849.469.82.622618.1 % 31.4 % 1,1913,643111409
ADL-Rundle-113.635.674.06.43215.6 % 25.0 % 3,2074,77356103
ADL-Rundle-338.439.278.30.5446.8 % 25.0 % 3025,9115260
KITTI-1638.349.971.61.0170.0 % 17.6 % 2198012960
KITTI-1933.847.968.21.0628.1 % 21.0 % 1,0592,41760191
Venice-131.835.772.91.21711.8 % 29.4 % 5392,5333870

Raw data:


QuadMOT
JAM