OVBT: Tracking Multiple Persons Based on a Variational Bayesian Model


Video not available.

Benchmark:

MOT16 |

Short name:

OVBT

Detector:

Public

Description:

An online variational
Bayesian model for multiple-person tracking is proposed. This yields a
variational expectation-maximization (VEM) algorithm. The computa-
tional efficiency of the proposed method is due to closed-form expressions
for both the posterior distributions of the latent variables and for the es-
timation of the model parameters. A stochastic process that handles
person birth and person death enables the tracker to handle a varying
number of persons over long periods of time.

Project page / code:

n/a

Reference:

Y. Ban, S. Ba, X. Alameda-Pineda, R. Horaud. Tracking Multiple Persons Based on a Variational Bayesian Model. In BMTT 2016, .

Processing:

Online

Last submitted:

August 16, 2016 (3 years ago)

Published:

July 16, 2016 at 12:07:44 CET

Submissions:

3

Open source:

No

Hardware:

2.53GHz

Runtime:

0.3 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
MOT1638.437.875.457 (7.5)359 (47.3)11,51799,46345.487.81.91,321 (29.1)2,140 (47.1)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
MOT16-0123.936.171.4396964,13735.376.41.53589
MOT16-0346.941.275.726306,17348,63153.590.14.16891,184
MOT16-0632.739.473.281295627,07338.788.80.5124183
MOT16-0733.634.273.35191,0779,60541.286.22.2158272
MOT16-0824.627.278.42261,06611,40231.983.31.7150177
MOT16-1232.840.076.79457664,74942.782.20.96380
MOT16-1418.126.874.541011,17713,86625.079.71.6102155

Raw data: