OVBT: Tracking Multiple Persons Based on a Variational Bayesian Model

MOT16-07


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.

Reference:

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

Last submitted:

August 16, 2016 (5 years ago)

Published:

July 16, 2016 at 12:07:44 CET

Submissions:

3

Project page / code:

n/a

Open source:

No

Hardware:

2.53GHz

Runtime:

0.3 Hz

Benchmark performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT1638.437.829.957 (7.5)359 (47.3)11,51799,46345.487.827.233.228.574.335.869.178.81.91,321 (29.1)2,140 (47.1)

Detailed performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT16-0123.936.128.1396964,13735.376.433.523.836.866.726.657.675.81.53589
MOT16-0346.941.232.626306,17348,63153.590.127.538.928.674.442.070.779.24.16891,184
MOT16-0632.739.429.781295627,07338.788.831.228.633.771.830.269.276.90.5124183
MOT16-0733.634.226.45191,0779,60541.286.223.530.224.474.132.367.676.62.2158272
MOT16-0824.627.222.72261,06611,40231.983.321.224.622.473.526.469.180.81.7150177
MOT16-1232.840.031.79457664,74942.782.233.030.734.977.434.065.579.80.96380
MOT16-1418.126.820.741011,17713,86625.079.723.418.524.475.219.863.077.31.6102155

Raw data: