We describe a real time authoring tool that is useful to supervise group of users in an augmented reality environment. The overall system is composed of a server; the Command Center (CC) and several clients; Wearable Systems (WS). The server is used to visualize the virtual representation of the real environment, 3d features map where augmentation are to occur, augmented content that will be rendered and users avatar that evolves on the field. The user's WS is equipped of a Head Mounted Display (HMD) along with sensors (GPS/DRM/Compass) to provide constant pose estimation (position and orientation) to the CC and a vision based tracking that uses natural features and pose estimation allgorithms to render virtual content accordingly. Alternatively a sensor based renderer is used to display AR guidance on the HMD based on the path defined and sent from the CC. The natural features system handles a large number of small 3D features maps. The overall system offers a complete system to train and supervise remote users through an AR environment.

This paper presents a new locally refined collision detection approach for large scale complex meshes in distributed virtual environments (DVEs) where exact and interactive interference detection is required. Transmitting models with millions of polygons is time consuming in comparison with transmitting simple models. Even if the models are transmitted in progressive manner, the earlier received models are in low level-of-detail (LOD) because the models are refined globally. Increasing the accuracy of collision detection (CD) at the client still takes time because increasing the LOD of the model is a slow process. The new approach is composed of an AB-tree collision query algorithm and a new mesh refinement algorithm on a space partitioned mesh (SPM) representation. It deals with this problem by selectively refining the models at certain areas that are predicted to collide with other objects and transmitting the refined parts instead of the entire model from server to client. The accuracy of CD is increased quickly at the predicted contact areas. An interactive rate is guaranteed by reducing network response time and dynamically adjusting the complexity and the space cost of the collision query algorithm

This paper presents a system for introducing augmented reality (AR) enhancements into an image-based cubic panorama sequence. Panoramic cameras, such as the Point Gray Research Ladybug allow rapid capture and generation of panoramic sequences for users to navigate and view. Our AR system provides the ability for authors to add virtual content into the panoramic sequences. First, a user manually selects a planar region over which to add the content. Then the system automatically finds a matching planar region in all the panoramas, allowing the virtual content to propagate. No preconditioning of the imaged scene through the addition of physical markers is necessary. Instead, 3-D position information is obtained by matching interest-point features across the panoramic sequence. This paper presents an application of augmented reality algorithms to the unique case of pre-captured panoramic sequences.

A robotic system is presented which automatically pots (i.e., sinks) pool balls. A homography is estimated that relates the gantry robot coordinate frame to the overhead (global) camera coordinate frame. This homography is computed by first calculating the mapping between the camera frame and a projection of the robot frame, and then solving the pool table plane equation in the robot frame. A measurement technique has been developed which is based upon a local camera attached to the robot end effector. This local camera allows the robot to be positioned accurately over circular targets placed on the table. The homography and table plane equation are then estimated by establishing correspondences between at least 4 measured target positions in the global camera and robot frames. The resulting homography allows the gantry to be positioned to within an average of 0.6 mm of a global camera frame position over the extent of a full sized pool table. The system has been used to pot a ball with 67% accuracy over the extent of the table, with a high repeatability.

We propose a method to augment live video based on the tracking of natural features, and the online estimation of the trinocular geometry. Previous without-marker approaches require the computation of camera pose to render virtual objects. The strength of our proposed method is that it doesn 7 require tracking of camera pose, and exploits the usual advantages of marker-based approaches for a fast implementation. A 3-view AR system is used to demonstrate our approach. It consists of an uncalibrated camera that moves freely inside the scene of interest, and of three reference frames taken at the time of system initialization. As the camera is moving, image features taken from an initial triplet set are tracked throughout the video sequence. And the trifocal tensor associated with each frame is estimated online. With this tensor, the square pattern that was visible in the reference frames is transferred to the video. This invisible pattern is then used by the ARToolkit to embed virtual objects.

Pattern-based augmented reality systems are considered the most promising approach for accurately registering virtual objects with real-time video feeds. The problem with existing solutions is the lack of robustness to partial occlusions of the pattern, which is important when attempting natural interactions with virtual objects. This paper describes a fast and accurate vision-based pattern tracking system that allows for autocalibrated 3D augmentation of virtual objects onto known planar patterns. The tracking system is shown to be robust to changes in pattern scale, orientation and, most importantly, partial occlusions. A method to detect a hand on top of the pattern is then described, along with a method to render the hand on top of the virtual objects.

Extracting geometric primitives from geometric sensor data is an important problem in model-based vision. A minimal subset is the smallest number of points necessary to define a unique instance of a geometric primitive. A genetic algorithm based on a minimal subset representation is used to perform primitive extraction. It is shown that the genetic approach is an improvement over random search and is capable of extracting more complex primitives than the Hough transform.>

A genetic algorithm based on a minimal subset representation of a geometric primitive is used to perform primitive extraction. A genetic algorithm is an optimization method that uses the metaphor of evolution, and a minimal subset is the smallest number of points necessary to define a unique instance of a geometric primitive. The approach is capable of extracting more complex primitives than the Hough transform. While similar to a hierarchical merging algorithm, it does not suffer from the problem of premature commitment.>

The problem of geometric primitive extraction in a range image is considered. A segmentation algorithm segments the image into quadric patches and provides an efficient initialization for the extraction of geometric primitives by a genetic algorithm. These algorithms are shown to be simple, fast and robust to noise. A parallel implementation has been realized and results are shown on real data.>

The problem of segmenting an image provided by a geometric sensor into geometric primitives is addressed by a two-step iterative process. In the first step, the largest connected region bounded by edge pixels is hypothesized as containing a geometric primitive. In the second step, the resulting set of pixels is sent to a robust fitter based on the least median of squares algorithm, to verify whether this hypothesis holds. If the fit is successful, then the geometric primitive is removed from the data; otherwise, a different hypothesis is obtained by decreasing the edge threshold. The process is repeated until the entire image is segmented. The fact that the robust fitter is tolerant of outliers means that the correct segmentation is produced from many different initial hypotheses. The algorithm is demonstrated on a number of range images which are known to contain particular geometric primitives.>

The acquisition of parts by a robot is an important problem industrially. The parts may be of various types and can be separate or jumbled together. We have mounted a laser rangefinder on the wrist of a robot and use the profile data it gives to determine a location, called a holdsite, at which to grasp the part. The laser rangefinder is lightweight and collects data quickly with more than sufficient accuracy for this application. The parts are not recognized, instead the best holdsite according to the criteria of stability and safety is chosen using methods from the field of spatial planning. The method has been tested on a variety of parts, from screws to rings and works well. The time required to acquire a part is approximately one minute but we believe this time can be decreased substantially to about ten seconds per part.