Energy Minimization by E � ective Jump Di � usion Method for Range Segmentation

This paper presents a stochastic jump di usion method for optimizing a Bayesian pos terior probability in segmenting range data and their associated re ectance images The algorithm works well on complex real world scenes indoor and outdoor which consist of an unknown number of objects or surfaces of various sizes and types such as planes conics smooth surfaces and cluttered objects like trees and bushes Formulated in the Bayesian framework the posterior probability is distributed over a countable number of subspaces of varying dimensions To search for globally optimal solution the paper adopts a stochastic jump di usion process to simulate a Markov chain random walk for exploring this com plex solution space A number of reversible jump dynamics realize the moves between di erent subspaces such as switching surface models and changing the number of objects The stochastic Langevin equation realizes di usions such as region competition in each subspace To achieve e ective computation the algorithm pre computes some importance proposal probabilities through Hough transforms edge detection and data clustering The latter is used by the Markov chains for fast mixing For the varying sizes scales of objects in natural scenes the algorithm computes in a multi scale fashion The algorithm is rst tested against an ensemble of D simulated data for performance analysis Then the algo rithm is applied to three datasets of range images under the same parameter setting The results are satisfactory in comparison with manual segmentation