This paper presents a study which uses an early cognitive vision system for model-based pose estimation of an object in motion. The estimated poses are processed by a Kalman filter which makes robotic manipulation possible. Results from a simulated robot environment are given.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
The paper presents the experimental comparison of ECV based grasping methods. We focus on three methods which generate grasping hypotheses based on visual information. One method based on edge information and two other ones based on surface information. These methods were tested by two kinds of experiments: simple one and complex one. Experiments were conducted at the University of Southern Denmark in the Cognitive Vision Lab. The research reveals the complementary power of given grasping methods. Furthermore it shows which method is the best for each of the objects. Owing to that, it is possible to improve grasping by using surface as well as edge information.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.