Multi-Class Multi-Object Tracking using Changing Point Detection with Hierarchical Data Model

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

Short name:

MCMOT_HDM

Benchmark:

Description:

n/a

Hardware:

2.4GHz, 4Core

Detector:

Private

Processing:

Batch

Last submitted:

July 11, 2016 (2 years ago)

Published:

July 14, 2016 at 02:02:49 CET

Submissions:

1

Open source:

No

Project page / code:

n/a

Reference:

B. Lee, E. Erdenee, S. Jin, M. Nam, Y. Jung, P. Rhee. Multi-Class Multi-Object Tracking using Changing Point Detection. In BMTT, 2016.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
62.478.31.731.5 % 24.2 % 9,85557,2571,3941,3182.4GHz, 4CorePrivate
IDF1ID PrecisionID Recall
51.660.744.9

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
MOT16-0154.747.876.80.82343.5 % 21.7 % 3492,5034243
MOT16-0373.255.478.43.014850.0 % 8.8 % 4,50723,060487472
MOT16-0657.753.079.70.522135.7 % 30.8 % 5534,143183162
MOT16-0754.444.378.61.25422.2 % 13.0 % 6196,678144128
MOT16-0841.934.978.82.26319.0 % 15.9 % 1,3788,132216210
MOT16-1242.353.180.51.38624.4 % 33.7 % 1,1653,5596253
MOT16-1442.048.274.71.716418.9 % 31.7 % 1,2849,182260250

Raw data:


MCMOT_HDM