Augmented reality (AR) deals with the problem of dynamically and accurately align virtual objects with the real world. Among the used methods, vision-based techniques have advantages for AR applications, their registration can be very accurate, and there is no delay between the motion of real and virtual scenes. However, the downfall of these approaches is their high computational cost and lack of robustness. To address these shortcomings we propose a robust camera pose estimation method based on tracking calibrated fiducials in a known 3D environment, the camera location is dynamically computed by the Orthogonal Iteration Algorithm. Experimental results show the robustness and the effectiveness of our approach in the context of real-time AR tracking.
Humanoid robots are not easy to program, it's a challenging task, even if they come with an advanced high level programming interfaces, such as drag-and-drop applications, which are indeed less complex and save a lot of time, it still needs an experienced user to produce a high quality motion, especially for the natural human-like motions which are very hard to reproduce. This paper introduces a new technique to fully imitate the human body motion on a humanoid robot NAO in real time using the Kinect Depth Camera, OpenNI Framework and the Robot Operating System (ROS).
International audienceThe implementation of an interaction between a user and a Virtual Environment (VE) in Virtual Reality (VR) may use various techniques. However, in some cases, the interaction must be very precise and comfortable for the user. In this paper, we introduce a new selection and manipulation technique called FOLLOW-ME. The model associated to this technique splits the Virtual Environment into three zones in which a specificinteraction model is used: a free manipulation zone, a scaled manipulation zone and a precise manipulation zone. Each one of the three zones is characterized by a specific interaction granularity which defines the properties of the interaction in the concerned zone. This splitting is created in order to have both precision near the object to reach or to manipulate (scaled and precise manipulation zones) and to maintain a realistic and free interaction in the VE (free manipulation zone). In the precise manipulation zone, we use the concept of virtual guides in order to assist user. In this paper, we exhibit the formalization of the model associated to the FOLLOW-ME technique. Then, we both compare our technique with classical interaction techniques in the case of a task which consists in trying to reach an object in VE and give some insights about the conditions of usefulness of virtual guides in a selection task. Some preliminary experimental results are presented and discussed