Linear Programming on 2D coordinates

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

Short name:

LP2D

Benchmark:

Description:

The tracker builds a graphical model where nodes are the 2D detections. The solution is then found by computing the minimum cost-flow problem.

Hardware:

2.6 GHz, 16 Cores

Detector:

Public

Processing:

Batch

Last submitted:

October 27, 2014 (3 years ago)

Published:

October 27, 2014 at 20:01:43 CET

Submissions:

1

Open source:

Yes

Project page / code:

Reference:

MOT baseline: Linear programming on 2D image coordinates.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
19.871.22.06.7 % 41.2 % 11,58036,0451,6491,7122.6 GHz, 16 CoresPublic
IDF1ID PrecisionID Recall
0.00.00.0

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing49.50.074.10.21315.4 % 15.4 % 494594841
PETS09-S2L240.70.070.21.9429.5 % 16.7 % 8484,554319359
ETH-Jelmoli40.70.073.50.74515.6 % 26.7 % 3191,1444157
ETH-Linthescher16.90.076.40.11972.0 % 73.6 % 1247,2227781
ETH-Crossing21.40.076.30.1263.8 % 65.4 % 197591010
AVG-TownCentre15.50.068.53.42268.4 % 33.2 % 1,5254,257260367
ADL-Rundle-12.90.072.27.73215.6 % 21.9 % 3,8534,934252207
ADL-Rundle-313.70.072.83.9442.3 % 25.0 % 2,4615,914400232
KITTI-1635.50.072.00.9170.0 % 11.8 % 1978534772
KITTI-1920.10.065.21.3628.1 % 21.0 % 1,3782,79397192
Venice-111.00.072.41.8170.0 % 35.3 % 8073,1569894

Raw data:


TBD
LP2D