DEPTH RECOVERY USING MARKOV RANDOM-FIELDS

Abstract

This paper deals with the problem of depth recovery or surface reconstruction from sparse and noisy range data. The image is modelled as a Markov Random Field and a new potential function is developed to effectively detect discontinuities in highly sparse and noisy images. No use of any line process is made. Interpolation over missing data sites is first done using local characteristics and simulated annealing is then used to compute the maximum a posteriori (MAP) estimate. Results of software simulations carried out on actual range images along with details of the simulations are presented.