Multi Object Tracking using Deep Structural Cost Minimization in Data Association


Short name:

UN_DAM

Benchmark:

Description:

n/a

Hardware:

3 GHZ, 1 Core GTx 1080

Detector:

Public

Processing:

Online

Last submitted:

June 09, 2019 (5 months ago)

Published:

June 14, 2019 at 13:41:22 CET

Submissions:

2

Open source:

No

Project page / code:

n/a

Reference:

Multi Object Tracking using Deep Structural Cost Minimization in Data Association

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
29.771.41.39.2 % 49.9 % 7,61035,2693186743 GHZ, 1 Core GTx 1080Public
IDF1ID PrecisionID Recall
41.458.332.1

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing60.368.675.10.31338.5 % 23.1 % 6636757
PETS09-S2L248.340.269.21.44214.3 % 14.3 % 6214,256111167
ETH-Jelmoli39.056.972.40.74517.8 % 35.6 % 3061,234839
ETH-Linthescher18.927.874.00.11973.6 % 75.6 % 1177,1141421
ETH-Crossing29.344.174.20.1267.7 % 61.5 % 1569223
AVG-TownCentre23.144.870.12.02269.7 % 51.3 % 8914,57727134
ADL-Rundle-119.342.772.55.93228.1 % 31.3 % 2,9424,5274165
ADL-Rundle-335.039.774.11.5446.8 % 34.1 % 9415,6204946
KITTI-1633.248.173.40.6170.0 % 29.4 % 1201,0051115
KITTI-1924.745.964.80.9623.2 % 29.0 % 8973,09431155
Venice-123.435.872.11.51711.8 % 35.3 % 6942,7831922

Raw data:

n/a


UN_DAM