Reply to "Comments on 'Joint Detection and Estimation of Multiple Objects from Image Observations'"
Ba‐Ngu Vo, Ba-Tuong Vo, Nam Trung Pham, David Suter · IEEE Transactions on Signal Processing · 2011
In this paper we present three theoretical results on conjugate priors for point processes (or random finite sets), namely Poisson, i.i.d. cluster, and multi-Bernoulli. As an example of the use of these results, the multi-Bernoulli conjugate prior was applied to multi-object filtering for image data, which covers various tracking problems, including track before-detect (TkBD). Davey's comments only concern our implemen tation of the HPMHT algorithm used to benchmark the performance of the multi-Bernoulli filter in a numerical example involving TkBD.