A Bayesian Framework for the Integration of Visual Modules

The Bayesian approach to vision provides a fruitful theoretical framework both for modeling individual cues, such as stereo, shading, texture and occlusion, and for integrating their information. In this formalism we represent the viewed scene by one, or more, surfaces using prior assumptions about the surface shapes and material properties. On theoretical grounds, the less information available to the cues (and the less accurate it is) then the more important these assumptions become. This suggests that visual illusions, and biased perceptions, will arise for scenes for which the prior assumptions are not appropriate. We describe psychophysical experiments which are consistent with these ideas. Our Bayesian approach also has two important implications for coupling diierent visual cues. Firstly, diierent cues cannot in general be treated independently and then simply combined together at the end. There are dependencies between them that have to be incorporated into the models. Secondly, a single generic prior assumption is not suucient even if it does incorporate cue interactions because there are many diierent types of visual scenes and diierent models are appropriate for each. This leads to the concept of competitive priors where the visual system must choose the corect model depending on the stimulus.