Search

International audience; This paper describes the ongoing developments in Photogrammetry and Mixed Reality for the Venus European project (Virtual ExploratioN of Underwater Sites, http://www.venus-project.eu). The main goal of the project is to provide archaeologists and the general public with virtual and augmented reality tools for exploring and studying deep underwater archaeological sites out of reach of divers. These sites have to be reconstructed in terms of environment (seabed) and content (artifacts) by performing bathymetric and photogrammetric surveys on the real site and matching points between geolocalized pictures. The base idea behind using Mixed Reality techniques is to offer archaeologists and general public new insights on the reconstructed archaeological sites allowing archaeologists to study directly from within the virtual site and allowing the general public to immersively explore a realistic reconstruction of the sites. Both activities are based on the same VR engine but drastically differ in the way they present information. General public activities emphasize the visually and auditory realistic aspect of the reconstruction while archaeologists activities emphasize functional aspects focused on the cargo study rather than realism which leads to the development of two parallel VR demonstrators. This paper will focus on several key points developed for the reconstruction process as well as both VR demonstrators (archaeological and general public) issues. The first developed key point concerns the densification of seabed points obtained through photogrammetry in order to obtain high quality terrain reproduction. The second point concerns the development of the Virtual and Augmented Reality (VR/AR) demonstrators for archaeologists designed to exploit the results of the photogrammetric reconstruction. And the third point concerns the development of the VR demonstrator for general public aimed at creating awareness of both the artifacts that were found and of the process with which they were discovered by recreating the dive process from ship to seabed.

International audience; In mobile outdoor augmented reality applications, accurate localization is critical to register virtual augmentations over a real scene. Vision-based approaches provide accurate localization estimates but still too sensitive to environmental conditions (change in brightness, sudden motion, occlusions, etc). This main drawback can be overcome by adding other types of sensors. In this paper, we combine a camera with an inertial sensor to estimate continuously the camera's orientation. The accuracy of this hybrid sensor depends on the accuracy of the calibration process that determines the relationship between the two sensors. This relationship allows expressing the orientation provided by the inertial sensor with respect to the camera coordinate system. The proposed method is simple, fast and accurate. Experimental results on real data are presented.

National audience

International audience; To properly align objects in the real and virtual worlds in an augmented reality (AR) space it is essential to keep tracking the camera's exact three-dimensional position and orientation (camera pose). State-of-the-art analysis shows that traditional vision-based or inertial sensor-based solutions are not adequate when used individually. Sensor fusion for hybrid tracking has become an active research direction during the past few years, although how to do it in a robust and principled way is still an open problem. In this paper, we develop a hybrid camera pose-tracking system that combines vision and inertial sensor technologies. We propose to use the particle filter framework for the sensor fusion system. Particle filters are sequential Monte-Carlo methods based upon a point mass (or 'particle') representation of probability densities, which can be applied to any state space model and which generalize the traditional Kalman filtering methods. We have tested our algorithm to evaluate its performance and have compared the results obtained by the particle filter with those given by a classical extended Kalman filter. Experimental results are presented.

National audience; Au cours de ces dernières années est apparue une nouvelle technologie, celle des Environnements Virtuels Collaboratifs (EVCs) qui n'est rien d'autre que le résultat de la convergence des intérêts de deux communautés de chercheurs : la Réalité Virtuelle (RV) et le Travail Collaboratif Assisté par Ordinateur. Les EVCs sont des systèmes qui peuvent supporter différentes formes de collaboration et d'interaction entre les utilisateurs. La collaboration dans ces systèmes se réfère à l'interaction simultanée (interaction collaborative IC) de plusieurs utilisateurs sur des objets virtuels. Même s'il existe plusieurs travaux dans ce domaine, il y a peu de chance de trouver des modèles qui contrôlent l'interaction et coordonne les actions des utilisateurs au sein des environnements virtuels monoutilisateur et multi-utilisateurs. Pour pallier ce problème, nous proposons dans cet article un forkflow pour l'interaction 3D collaborative afin de gérer l'interaction de groupe et les échanges d'information au sein d'un environnement virtuel immersif ou semi-immersif. Le formalism de l'interaction 3D est basé sur l'awareness de groupe (focus, nimbus et degrée d'interaction) combiné astucieusement avec les différents paradigmes de l'interaction 3D (la navigation, la sélection et la manipulation). Nous présentons aussi le workflow qui représente un cadre qui permet de structurer et de gérer l'interaction simultnée des différentes utilisateurs qui composent l'environnement virtuel collaboratif.

National audience; In this paper, we will introduce a new componentbased framework for mixed and augmented reality applications. We will see the developed framework intends to meet several requirements such as portability, variable component granularity, scalability, and a high level of abstraction for end-users. The provided framework, at runtime, is composed of a set of several libraries where components are stored, an XML file in which configuration and communications between components are described and a runtime program that deals with the two previous parts of the system. Our component model can also deal with reconfigurable data-flows by the use of a finite state-machine, using several component configurations within an application.

National audience; In this paper we present a visual object tracking approach to extract motion information for Augmented Reality (AR) systems. Our proposed system tracks the target object by applying a model based pose estimationalgorithm. The approach is to fuse information from an active contours model and optical flow motion estimation.The optical flow is used to provide a constraint on the deformable model motion and place the initialcontour in the region of interest of the active contour. For pose estimation we use the Extended Kalman Filter(EKF), the measurement equation models the feature points of object in image and the process model predictsthe behavior of the system based on the current state and estimates the position and orientation of the objecttoward the camera coordinate frame. The algorithm is tested in real time and shows to be robust and efficient.

International audience; In this paper, we present a robust fiducials tracking method for real time Augmented Reality systems. We use a generic algorithm for object detection and feature points extraction, then, we identify the target object with its barcode for template matching. Given a 2D feature points and a 3D object model, object pose consists in recovering the position and the orientation of the object with respect to the camera. Using the computed pose and the identified object, we propose a first tracking algorithm based on homography computation. However due to lighting or contrast conditions or occlusion of the target object by an other object, the tracking may fail. Therefore, we extend the previous object pose algorithm for bar-coded fiducials, using a second algorithm based on RANSAC estimator and fiducial reprojection on the image. The proposed method is computationally efficient and showed to be accurate and robust.

International audience; This paper presents a robust vision-based solution for camera pose tracking that can handle large camera displacements, drastic aspect changes, and partial occlusions. It is based on tracking calibrated fiducials. A RANSAC approach is used for robust matching in the presence of outliers to deal with partial occlusions. The camera location is computed dynamically using the Orthogonal Iteration (OI) Algorithm. Some experimental results are presented and discussed.

International audience; The AMRA project, carried out by a consortium including industrials and research partners, aims at implementing an Augmented Reality (AR) system for mobile use in industrial applications such as train maintenance and repairs in industrial sites. The adopted solution is a video see-through system where a tablet-PC is used as an augmented window. The overall architecture of a prototype is unfolded, and its key points are detailed. For instance, a visual registration system has been developed to accurately overlay a video stream with information. A robust, real time registration, using a single camera tied to the tablet-PC, is performed. Besides, a hierarchical description of maintenance procedure is set up and enriched by new media such as photos, video and/or 3D models. These 3D models have been specially tailored to meet maintenance tasks requirements. The obtained multimedia contents allow easy access to technical documentation through a man machine interface managing a multimedia engine. All these features have been combined in the AMRA prototype which have been evaluated by a maintenance operator.

International audience; Augmented Reality (AR) can provide to a Human Operator (HO) a real help to achieve complex tasks such as remote control of robots and cooperative teleassistance. Using appropriate augmentations, the HO can interact faster, safer and easier with the remote real world. This paper presents a new Man Machine Interface (MMI) allowing high-level telewok, using augmented reality technologies. An example of telework is presented in this paper as a teleoperation of the remote 4 Degree of Freedom (DoF) robot. An important feature of this interface is based on the use of two new modes devoted for HO perception and interaction.In a new perception mode, a wide screen with stereo glasses are used to allow stereoscopic perception of the remote real environment. The new interaction mode is given thanks to ART optical tracking system that allows HO to interact freely with the remote 4 DoF robot.An other important work presented in this paper is a distributed software architecture developed with three client/server modules (robot control, stereoscopic video feedback and tracking) and using appropriate methods (communication protocols, image compressions etc.).This new interface is tested and compared with ARITI interface (acronym of Augmented Reality Interface for Teleoperation via Internet) and some preliminary results are presented and discussed.

International audience; High accuracy registration between real and virtual environments is crucial in Augmented Reality (AR) systems. However, when a vision/inertial hybrid tracker is used,such accuracy depends mostly on the calibration procedure to determine transformations between the sensors frames. This calibration allows to project all data in a single reference frame. In this paper, we describe a new calibration method for a hybrid tracking system. It consists on rigidly assembling the hybrid tracker to a 6DOF robot in order to simulate the users head motion while tracking targets in AR environment. Our approach exploits the robot positioning to obtain a high accuracy for the tracker calibration. Experimental results and accuracy analyses are presented and demonstrate our approach effectiveness.