Towards a Constraint Based Theory of Sensory Data Fusion
We have observed that information acquisition is required by organisms in order to determine the state of the environment that they are operating in. What information is required is related to the activities that the organism undertakes. In order to obtain the required world information one must invert the world-image mapping. In general this mapping is non-invertible so that extra information must be added to constrain the space of world configurations sufficiently to allow a unique solution to be obtained.
KeywordsData Fusion Markov Random Field Prior Constraint Prior Model Smoothness Constraint
Unable to display preview. Download preview PDF.