Efficient Monte Carlo Sampler for Detecting Parametric Objects in Large Scenes

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.

[1]  Florent Lafarge,et al.  Geometric Feature Extraction by a Multimarked Point Process , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Robert T. Collins,et al.  Marked point processes for crowd counting , 2009, CVPR.

[3]  Josiane Zerubia,et al.  Higher Order Active Contours , 2006, International Journal of Computer Vision.

[4]  M.N.M. vanLieshout.  Depth map calculation for a variable number of moving objects using Markov sequential object processes , 2007 .

[5]  Xavier Descombes,et al.  Author manuscript, published in "Journal of Mathematical Imaging and Vision (2009)" Object , 2008 .

[6]  Arthur Gretton,et al.  Parallel Gibbs Sampling: From Colored Fields to Thin Junction Trees , 2011, AISTATS.

[7]  P. Green,et al.  Parallel Chains, Delayed Rejection and Reversible Jump MCMC for Object Recognition , 2000, BMVC.

[8]  U. Grenander,et al.  Jump–diffusion Markov processes on orthogonal groups for object pose estimation , 2002 .

[9]  Josiane Zerubia,et al.  Point processes for unsupervised line network extraction in remote sensing , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  Tianxu Zhang,et al.  Marked Point Process for Vascular Tree Extraction on Angiogram , 2007, EMMCVPR.

[11]  Zhuowen Tu,et al.  Range image segmentation by an effective jump-diffusion method , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .

[13]  Stephen A. Jarvis,et al.  On the parallelisation of MCMC-based image processing , 2010, 2010 IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum (IPDPSW).

[14]  Ákos Utasi,et al.  A 3-D marked point process model for multi-view people detection , 2011, CVPR 2011.

[15]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[16]  A. Baddeley,et al.  Stochastic geometry models in high-level vision , 1993 .

[17]  Zhuowen Tu,et al.  Image Segmentation by Data-Driven Markov Chain Monte Carlo , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  Florent Lafarge,et al.  Towards the parallelization of Reversible Jump Markov Chain Monte Carlo algorithms for vision problems , 2012 .

[19]  Uwe Soergel,et al.  A Marked Point Process for Modeling Lidar Waveforms , 2010, IEEE Transactions on Image Processing.

[20]  Uwe Soergel,et al.  Lidar waveform modeling using a marked point process , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).