Concurrent Stereo Reconstruction

This thesis presents a probabilistic approach to multi view stereo reconstruction from calibrated images. Together with a voxel based world model a Markov Random Field is used to describe the problem of stereo reconstruction as a structural recognition task. A speci c scene representation has been developed for stereo reconstruction with the ability to keep ambiguities within the reconstruction result for further evaluation. Based on several implementation speci c methods being used in combination with Gibbs sampling an e ciently computable and exible framework for stereo reconstruction is presented to solve the Bayesian estimation task. Di erent a priori and observation models are investigated to adapt the general model to speci c reconstruction tasks and unsupervised learning methods have been used to reduce the number of model parameters. Evaluation on su cient test data proves the viability of the proposed method.