Control of Perception in Dynamic Vision

In order to achieve continuous operation and thus facilitate use of vision in a dynamic scenario it is necessary to introduce a purpose for the visual processing in order to provide information that may control the visual processing and thus limit the amount of resources needed to obtain the required results. A proposed architecture for vision systems is presented together with an architecture for visual modules. This architecture enables both goal and data driven processing with a potentially changing balance between the two modes. To illustrate the potential of the proposed architecture an example system for recovery of scene depth is presented together with experimental results which demonstrates a scalable performance.