K-Dense Neighbors based Tracker

MOT16-01 MOT16-03 MOT16-06 MOT16-07 MOT16-08 MOT16-12 MOT16-14

Short name:

KDNT

Benchmark:

Description:

Reid Feature from :
Liu Y, Yan J, Ouyang W. Quality Aware Network for Set to Set Recognition[J]. arXiv preprint arXiv:1704.03373, 2017.

Hardware:

2.8 GHZ, 1 Core

Detector:

Private

Processing:

Batch

Last submitted:

August 02, 2017 (1 year ago)

Published:

September 20, 2016 at 02:43:09 CET

Submissions:

1

Open source:

No

Project page / code:

n/a

Reference:

F. Yu, W. Li, Q. Li, Y. Liu, X. Shi, J. Yan. POI: Multiple Object Tracking with High Performance Detection and Appearance Feature. In BMTT, SenseTime Group Limited, 2016.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
68.279.41.941.0 % 19.0 % 11,47945,6059331,0932.8 GHZ, 1 CorePrivate
IDF1ID PrecisionID Recall
60.066.954.4

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
MOT16-0151.762.778.12.62356.5 % 4.3 % 1,1481,8974246
MOT16-0381.464.879.43.814875.7 % 3.4 % 5,63113,539267253
MOT16-0663.659.580.80.622143.9 % 19.5 % 6953,41098124
MOT16-0758.454.480.81.65440.7 % 7.4 % 8055,88799113
MOT16-0839.645.381.40.96322.2 % 33.3 % 5669,44896112
MOT16-1248.257.580.10.58620.9 % 39.5 % 4573,7954653
MOT16-1445.445.975.82.916421.3 % 22.0 % 2,1777,629285392

Raw data:


KDNT