Skip to Main content Skip to Navigation
Reports

Towards the parallelization of Reversible Jump Markov Chain Monte Carlo algorithms for vision problems

Yannick Verdie 1, 2 Florent Lafarge 1
1 GEOMETRICA - Geometric computing
CRISAM - Inria Sophia Antipolis - Méditerranée , Inria Saclay - Ile de France
Abstract : Point processes have demonstrated efficiency and competitiveness when addressing object recognition problems in vision. However, simulating these mathematical models is a difficult task, especially on large scenes. Existing samplers suffer from average performances in terms of computation time and stability. We propose a new sampling procedure based on a Monte Carlo formalism. Our algorithm exploits Markovian properties of point processes to perform the sampling in parallel. This procedure is embedded into a data-driven mechanism such that the points are non-uniformly distributed in the scene. The performances of the sampler are analyzed through a set of experiments on various object recognition problems from large scenes, and through comparisons to the existing algorithms.
Document type :
Reports
Complete list of metadatas

https://hal.inria.fr/hal-00720005
Contributor : Florent Lafarge <>
Submitted on : Monday, July 23, 2012 - 1:11:17 PM
Last modification on : Saturday, January 27, 2018 - 1:31:38 AM
Long-term archiving on: : Wednesday, October 24, 2012 - 2:35:56 AM

File

RR-8016.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00720005, version 1

Collections

Citation

Yannick Verdie, Florent Lafarge. Towards the parallelization of Reversible Jump Markov Chain Monte Carlo algorithms for vision problems. [Research Report] RR-8016, INRIA. 2012. ⟨hal-00720005⟩

Share

Metrics

Record views

391

Files downloads

185