Abstract

Computer vision is becoming an integral part in human-machine interfaces as research increasingly aims at a seamless
and natural interaction between a user and an application system. Gesture recognition, context awareness, and grounding
concepts in the commonly perceived environment as well as in the interaction history are key abilities of such systems.
Simultaneously, recent computer vision research has indicated that integrated systems which are embedded in the world
and interact with their environment seem a prerequisite for solving more general vision tasks. Cognitive computer vision
systems which enable the generation of knowledge on the basis of perception, reasoning, and extension of prior models
are a major step towards this goal. For these, the integration, interaction and organization of memory becomes a key
issue in system design. In this article we will present a computational framework for integrated vision systems that is
centered around an active memory component. It supports a fast integration and substitution of system components,
various means of interaction patterns, and enables a system to reason about its own memory content. This framework
will be exemplified by means of a cognitive human-machine interface in an Augmented Reality scenario. The system is
able to acquire new concepts from interaction and provides a context aware scene augmentation for the user.