View-based detection of 3-D interaction between hands and real objects

We propose a vision-based method to detect interactions between human hand(s) and real objects. Since humans perform various kinds of tasks with their hands, detection of hand-object interactions is useful for building intelligent systems that understand and support human activities. We use a statistical color model to detect hand regions in input images. Target objects are dynamically modeled based on their appearances by giving consideration to occlusions by the hand. The appearance model tracks the translation and relative rotation of target objects. This system is useful for recording, indexing and instructing object manipulations and/or hand-object interactions. Experimental results show the effectiveness of our method.