Gaussian mixture PHD and CPHD filtering with partially uniform target birth
MetadataShow full item record
The standard Gaussian Mixture Probability Hypothesis Density (GMPHD) filter and Cardinalised Probability Hypothesis Density (GMCPHD) filter require the target birth model to take the form of a Gaussian mixture. Although any density (including a uniform density), can be approximated using a sum of Gaussians, this can be inefficient in practice, especially when a large number of Gaussians is required to achieve the desired accuracy. A better alternative in the case of an uninformative birth model would be to directly use a uniform density instead of a Gaussian mixture approximation. In this paper we present new forms of the GMPHD and GMCPHD filtering equations, which allow part of the target birth model to take on a uniform distribution, thus obviating the need to use large Gaussian mixtures to approximate a uniform birth density.
Showing items related by title, author, creator and subject.
Ma, L.; Wang, P.; Xue, K.; Kim, Du Yong (2014)© 2014 IEEE. Recently, the Gaussian Mixture Probability Hypothesis Density (GMPHD) filter has been studied as a popular method for multi-target tracking in clutter background. As an extended version, robust GMPHD filter, ...
Beard, Michael; Vo, Ba; Vo, Ba-Ngu; Arulampalam, S. (2013)The conventional GMPHD/CPHD filters require the PHD for target births to be a Gaussian mixture (GM), which is potentially inefficient because careful selection of the mixture parameters may be required to ensure good ...
Yoon, J.; Kim, Du Yong; Bae, S.; Shin, V. (2011)In this correspondence, a new multi-target tracking (MTT) algorithm based on the probability hypothesis density (PHD) filtering framework is designed in order to improve tracking performance via the proposal of two ...