This paper proposes an automatic functional object segmentation method based on modeling the relationship between grasping hand form and the object appearance. First the relationship among a representative grasping pattern and a position and pose of a object relative to the hand is learned based on a few typical functional objects. By learning local features from the hand grasping various tools...

Sensor feedback and sensor fusion are indispensable for working in the real world. In this paper a robot hand grasping method is proposed. This method uses visual and tactile sensor information. First, grasping evaluation functions is proposed, which is derived from the relationships between a hand and an object. Then a control method is proposed which uses visual and tactile feedback. In this ...

Grasping unknown objects in natural environments remains an open problem. An important question is how we devise simple, fast and robust methods that rely on minimal representations of objects. In this report, we present a stereo system that generates grasping hypotheses based on curve matching between two views. The contribution of the work is the use of dynamic time warping approach for enfor...

Imitation learning holds the promise of robots which need not be programmed but instead can learn by observing a teacher. We present recent efforts being made at our laboratory towards endowing a robot with the capability of learning to imitate human hand gestures. In particular, we are interested in grasping movements. The aim is a robot that learns, e.g., to pick up a cup at its handle by imi...