GMPHD_15: Online Multi-Object Tracking with Hierarchical Data Association using the GMPHD filter


Benchmark:

Short name:

GMPHD_15

Detector:

Public

Description:

Stage 1: low-level association between targets states and observations(i.e., detection responses), frame by frame
-We only used detection reponses with the positive confidence value among provided detection results

Stage 2: short tracklet elimination under the set length(parameter)
-Minimum tracklet length for mid-level association : 10

Stage 3: mid-level association between rough tracklets(almost fragmented or ID swtiched)
-Only Motion and size are used
-Appearance is not used

Project page / code:

Reference:

Y. Song, M. Jeon. Online Multiple Object Tracking with the Hierarchically Adopted GM-PHD Filter using Motion and Appearance. In IEEE/IEIE The International Conference on Consumer Electronics (ICCE) Asia, 2016.

Processing:

Online

Last submitted:

December 05, 2016 (3 years ago)

Published:

December 05, 2016 at 10:21:45 CET

Submissions:

1

Open source:

No

Hardware:

3.5 GHz, 4 Cores

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN ID Sw.
2D MOT 201518.528.470.928.0399.07,86441,766459

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN ID Sw.
ADL-Rundle-110.028.472.12.013.02,3665,95458
ADL-Rundle-315.323.873.20.018.01,3747,17464
AVG-TownCentre16.328.470.09.0147.04135,54226
ETH-Crossing6.422.574.22.019.01807563
ETH-Jelmoli34.342.872.25.016.03171,33516
ETH-Linthescher16.224.072.25.0142.05396,89644
KITTI-1627.537.873.20.05.01541,06119
KITTI-1911.533.365.33.017.01,2853,39450
PETS09-S2L231.928.469.10.013.04675,965131
TUD-Crossing50.549.272.32.02.04148519
Venice-113.222.871.50.07.07283,20429

Raw data:


GMPHD_15