When reading this paper, a key thing to remember (mentioned in the introduction) is that the term "particle" refers to weighted samples which encode nonparametric probabilities (parzen-window style) in various subspaces.

A nice ICCV 2005 Shourt Course on MCMC by Zhu, Delleart and Tu

http://civs.stat.ucla.edu/MCMC/MCMC_tutorial.htm

When talking about DDMCMC I might also mention the Swendsen-Want Cut technique. If anybody is interested they should check out the ICCV 2003 paper http://www.stat.ucla.edu/~abarbu/Research/iccv2003.pdf

Pleas post you thoughts on "Image Segmentation by Data-Driven Markov Chain Monte Carlo" here.