A uniform Bayesian framework for integration

Vision researchers have advocated the integration of vision modules. However, generic system integration issues for recovering 3D information have not been adequately addressed in the literature. The authors propose a unified Bayesian integration framework for interactions among the vision modules to obtain a complete 3D reconstruction from a pair of intensity (stereo) images. The authors have integrated perceptual grouping, stereo, shape from shading, and shape from texture modules under the proposed framework. They have demonstrated that the integrated system recovers the depth and surface orientation information more reliably than the individual modules for different synthetic and real images.