Continuous Energy Minimization for Multi-target Tracking

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

Short name:

CEM_16

Benchmark:

Description:

Contrary to recent approaches, we focus on designing a continuous energy that corresponds
to a more complete representation of the problem, rather than one that
is amenable to global optimization. Besides the image evidence, the
energy function takes into account physical constraints, such as target
dynamics, mutual exclusion, and track persistence. In addition, partial
image evidence is handled with explicit occlusion reasoning, and
different targets are disambiguated with an appearance model. To
nevertheless find strong local minima of the proposed non-convex energy
we construct a suitable optimization scheme that alternates between
continuous conjugate gradient descent and discrete trans-dimensional
jump moves. These moves, which are executed such that they always reduce
the energy, allow the search to escape weak minima and explore a much
larger portion of the search space of varying dimensionality.

Hardware:

3 GHz, 1 CPU

Detector:

Public

Processing:

Batch

Last submitted:

March 22, 2016 (9 months ago)

Published:

March 21, 2016 at 00:00:00 CET

Submissions:

1

Open source:

No

Project page / code:

n/a

Reference:

A. Milan, S. Roth, K. Schindler. Continuous Energy Minimization for Multitarget Tracking. In IEEE TPAMI, 2014.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
33.275.81.27.8 % 54.4 % 6,837114,3226427313 GHz, 1 CPUPublic

Detailed performance:

Sequence MOTA MOTP FAF GT MT ML FP FN ID Sw Frag
MOT16-0128.472.30.32317.4 % 43.5 % 1234,4203631
MOT16-0338.375.92.214811.5 % 39.9 % 3,25161,110168238
MOT16-0633.974.20.42216.8 % 56.6 % 5276,960138143
MOT16-0728.374.11.6547.4 % 48.1 % 79310,808107126
MOT16-0826.780.01.0639.5 % 47.6 % 63211,5537774
MOT16-1234.977.90.78612.8 % 52.3 % 6514,7193435
MOT16-1415.173.41.11641.2 % 72.0 % 86014,7528284

Raw data:


CEM_16