LDCT: Learning to Divide and Conquer Tracker

AVG-TownCentre


Benchmark:

Short name:

LDCT

Detector:

Public

Description:

Test are performed on fixed camera video only:
- TUD-Crossing
- PETS09-S2L2
- AVG-TownCentre
- ADL-Rundle-3
- KITTI-16
- Venice-1

Results on other sequences are not meaningful.

Reference:

F. Solera, S. Calderara, R. Cucchiara. Learning to Divide and Conquer for Online Multi-Target Tracking. In ICCV, 2015

Last submitted:

April 22, 2015 (5 years ago)

Published:

April 22, 2015 at 22:31:13 CET

Submissions:

1

Project page / code:

Open source:

No

Hardware:

3 GHZ, 1 Core

Runtime:

20.7 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
2D MOT 20154.716.871.782 (11.4)234 (32.5)14,06632,15647.767.62.412,348 (259.1)2,918 (61.2)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
ADL-Rundle-1-68.00.370.8946,3584,14155.544.812.75,133570
ADL-Rundle-325.220.273.42274537,03930.887.30.7110101
AVG-TownCentre31.742.072.236241,8782,60863.570.74.2395509
ETH-Crossing-6.12.573.73107870629.679.20.428044
ETH-Jelmoli-23.71.672.88116411,09756.869.21.51,399182
ETH-Linthescher-1.72.073.541322686,74024.589.10.22,076435
KITTI-1653.043.479.0239166560.991.90.44457
KITTI-19-48.01.266.63142,6262,79247.749.32.52,490561
PETS09-S2L247.426.970.8639953,77960.885.52.3297300
TUD-Crossing67.752.582.99110020581.490.00.55130
Venice-133.524.768.4055782,38447.879.01.373129

Raw data: