Abstract We present a marker-less AR/DR system that can replace the arm of the user with a virtual bionic prosthesis in real time including finger tracking. For this, we use a mixed reality HMD that provides the user with a stereo image based on video-see-through (VST). We apply chroma-keying to remove the user’s arm from each captured image and input reconstructed background information into the removed pixels. Before rendering the prosthesis model into the image, we re-target motion capture data of the user’s hand to the kinematic skeleton of the prosthesis to match the current hand pose. This system opens new research possibilities on self- and other-perception of bionic bodies. In a first evaluation study of the system, we propose that users perceive the virtual prosthesis model as a part of their body (i.e., that they experience a sense of ownership). We tested this assumption in a laboratory study with 27 individuals who used the system to perform a series of simple tasks in AR with their prosthesis. We measured body ownership and other measures with self-reports. In support of the hypothesis, users experienced a sense of body ownership. Also, a feeling of self-presence is induced during the task, and participants rated the overall experience as positive.

Abstract We present in the paper a hybrid method for motion editing combining motion blending and Jacobian‐based inverse kinematics (IK). When the original constraints are changed, a blending‐based IK solver is first employed to find an adequate joint configuration coarsely. Using linear motion blending, this search corresponds to a gradient‐based minimization in the weight space. The found solution is then improved by a Jacobian‐based IK solver by further minimizing the distance between the end effectors and constraints. To accelerate the searching in the weight space, we introduce a weight map, which pre‐computes the good starting positions for the gradient‐based minimization. The advantages of our approach are threefold: first, more realistic motions can be generated by utilizing motion blending techniques, compared with pure Jacobian‐based IK. The blended results also increase the rate of convergence of the Jacobian‐based IK solver. Second, the Jacobian‐based IK solver modifies poses in the pose configuration space and the computational cost does not scale with the number of examples. Third, it is possible to extrapolate the given example motions with a Jacobian‐based IK solver, while it is generally difficult with pure blending‐based techniques. Copyright © 2014 John Wiley & Sons, Ltd.

When users are interacting collaboratively in a virtual environment it cannot be guaranteed that every user has the same input device or that they have access to the same information. Our research aims at understanding the effects of such asymmetries on the user embodiment in collaborative virtual environments (CVEs). In order to do this, we have developed a prototyping platform for cooperative interaction between two users. To change the information a user has, we are incorporating “special views”[1] for each person. Also, an easily expandable array of input devices is supported, e.g. Mouse/Keyboard, Novint Falcon, Razer Hydra, ART Flightstick, Leap Motion, etc. Those devices can provide additional information to a user, like haptic feedback, but they can also be used to restrain a user, for example by providing a device with less degrees of freedom.

When users are interacting in a collaborative virtual environment it can neither be guaranteed that every user has the same input device nor that they have access to the same information. Our research aims at understanding the effects of such asymmetries on the user embodiment in cooperative multi-user environments. In order to do this, a generic interaction application is needed, which can be altered quickly to generate differences in information distribution and interaction possibilities. In this paper we therefore present the development of such a prototyping platform for cooperative interaction between two users in a stereoscpoic cooperative virtual world. A flexible software framework enables us to quickly generate physics-based puzzles for the users to solve together. To change the information a user gets, we are incorporating "special views" for each person. To diversify interaction fidelity, an easily expandable array of input devices is supported, e.g. Mouse/Keyboard, Novint Falcon, Razer Hydra, ART Flight stick, etc. Those devices can provide additional information to a user, like hap tic feedback, but they can also restrain a user by providing less degrees of freedom. Additionally, they can be used to compare different interaction metaphors, like handover, object possession, or picking behavior.

Abstract Lossy compression of motion capture data can alleviate the problems of efficient storage and transmission by exploiting the redundancy and the superfluous precision of the data. When considering the acceptable amount of distortion, perceptual issues have to be taken into account. Current state of the art methods reduce the data rate required for high quality storage of motion capture data using various techniques. Most of them, however, do not use the common tools of general data compression, such as the method of Lagrange multipliers, and thus they obtain sub‐optimal results, making it difficult to do a fair comparison of their performance. In this paper, we present a general preprocessing step based on Lagrange multipliers, which allows to rigorously adjust the precision in each of the degrees of freedom of the input data according to the amount of influence the given degree of freedom has on the overall distortion. We then present a simple compression method based on Principal Component Analysis, which in combination with the proposed preprocessing achieves significantly better results than current state of the art methods. It allows optimization with respect to various distortion metrics, and we discuss the choice of the metric in two common but distinct scenarios, proposing a perceptually oriented comparison metric based on the relation of the problem at hand to the problem of compression of dynamic meshes.

Abstract In this paper, we investigate the possibilities of efficient encoding of UV coordinates associated with vertices of a triangle mesh. Since most parametrization schemes attempt to achieve at least some level of conformality, we exploit the similarity of the shapes of triangles in the mesh and in the parametrization. We propose two approaches building on this idea: first, applying a recently proposed generalization of the parallelogram predictor, using the inner angles of mesh triangles corresponding to the UV‐space triangles. Second, we propose an encoding method based on discrete Laplace operator, which also allows exploiting the information contained in the mesh geometry to efficiently encode the parametrization. Our experiments show that the proposed approach leads to savings of up to 3 bits per UV vertex, without loss of precision.

Abstract This paper addresses the problem of representing dynamic 3D meshes in a compact way, so that they can be stored and transmitted efficiently. We focus on sequences of triangle meshes with shared connectivity, avoiding the necessity of having a skinning structure. Our method first computes an average mesh of the whole sequence in edge shape space. A discrete geometric Laplacian of this average surface is then used to encode the coefficients that describe the trajectories of the mesh vertices. Optionally, a novel spatio‐temporal predictor may be applied to the trajectories to further improve the compression rate. We demonstrate that our approach outperforms the current state of the art in terms of low data rate at a given perceived distortion, as measured by the STED and KG error metrics.

Many algorithms have been proposed for the task of efficient compression of triangular meshes. Geometric properties of the input data are usually exploited to obtain an accurate prediction of the data at the decoder. Considerations on how to improve the prediction usually focus on its normal part, assuming that the tangential part behaves similarly. In this paper, we show that knowledge of vertex valences might allow the decoder to form a prediction that is more accurate in the tangential direction, using a weighted parallelogram prediction. This idea can be easily implemented into existing compression algorithms, such as Edgebreaker, and it can be applied at different levels of sophistication, from very simple ones, that are computationally very cheap, to some more complex ones that provide an even better compression efficiency.

Abstract This paper presents an algorithm for the real time rendering of fur without adding fur‐specific geometry, such as shells, to the object. It is based on Cone Step Mapping and introduces a local distortion of the view vector to simulate a deformation of the heightfield‐bound hair geometry. While this distortion enables a more realistic fur rendering, some limitations emerge and have to be dealt with. A local light reflectance model including approximations of global light interactions with hair and skin is proposed. We furthermore show how material and geometric properties can locally be influenced through standard texture mapping. This includes most notably the local modification of growth and streak direction of the hairs.

Abstract In this paper we present a novel method to approximate the force field of a discrete 3d object with a time complexity that is linear in the number of voxels. We define a rule, similar to the distance transform, to propagate forces associated with boundary points into the interior of the object. The result of this propagation depends on the order in which the points of the object are processed. Therefore we analyze how to obtain an order‐invariant approximation formula. With the resulting formula it becomes possible to approximate the force field and to use its features for a fast and topology preserving skeletonization. We use a thinning strategy on the body‐centered cubic lattice to compute the skeleton and ensure that critical points of the force field are not removed. This leads to improved skeletons with respect to the properties of centeredness and rotational invariance.

Abstract We present a reverse engineering method for constructing a surface approximation scheme whose input is a set of unorganized noisy points in space and whose output is a set of quadric patches. The local surface properties, necessary for the subsequent segmentation, are estimated directly from the data using a simple and efficient data structure—the neighborhood graph. Our segmentation scheme, based on principal curvatures, constructs initial point subsets, which may be enlarged or further subdivided based on associated approximation error estimates obtained through approximation of the initial segments by quadric surfaces. Our method is highly efficient and produces a high‐quality piecewise quadric surface approximation of engineering objects, which we demonstrate for several simple and complex example data sets.

In this paper, we report on the realization of an immersive table tennis simulation. After describing the hardware necessities of our system, we give insight into different aspects of the simulation. In particular, the developed methods for collision detection and physical simulation are presented. The design of the virtual opponent is of crucial importance to realize an enjoyable game. Therefore, we report on the implemented game strategy and the animation of the opponent. Since table tennis is one of the fastest sports, the synchronization of the human player's movements and the visual output on the projection wall is a very challenging problem to solve. To overcome the latencies in our system, we designed a prediction method that allows high speed interaction with our application.

In this paper we summarize the experiences we made with the implementation of a table tennis application. After describing the hardware necessities of our system we give insight into different aspects of the simulation. These include collision detection, physical simulation and some aspects of the design of the virtual opponent.Since table tennis is one of the fastest sports the synchronization of the player's movements and the visual output on the projection wall is the most challenging problem to solve. Therefore we analysed the latencies of all subcomponents of our system and designed a prediction method that allows high speed interaction with our application.

We present a novel LOD based rendering approach that integrates geometry based and image based LOD mechanisms in one system. We assume that every LOD object in the scene is represented by a textured polygonal mesh. In a preprocess we generate a continuous multiresolution model for all LOD objects. During runtime we first check for every LOD object if an appropriate cached approximation generated for former frames is available. If not, we extract a new suitable approximation out of the objects continuous multiresolution model depending on view-dependent criteria. After that, we replace this adaptive geometry-based approximation by an image-based impostor if this is reasonable and efficient. The selected approximations of LOD objects are used for rendering the next frame.

In reverse engineering a physical object is digitally reconstructed from a set of boundary points. In the segmentation phase these points are grouped into subsets to facilitate consecutive steps as surface fitting. In this paper we present a step segmentation method with subsequent classification of simple algebraic surfaces. Our method is direct in the sense that it operates directly on the point set in contrast to other approaches that are based on a triangulation of the data set. The segmentation process involves a fast algorithm for k-nearest neighbors search and an estimation of first and second order surface properties. First order segmentation, based on normal vectors, provides an initial subdivision of the surface and detects sharp edges as well as flat or highly curved areas. One of the main features of our method is to proceed by alternating the steps of segmentation and normal vector estimation. Second order segmentation subdivides the surface according to principal curvatures and provides a sufficient foundation for the classification of simple algebraic surfaces. If the boundary of the original object contains such surfaces the segmentation is optimized based on the result of a surface fitting procedure.

The problem of k-nearest neighbors computation within a 3D data set is frequently encountered in computer graphics. Applications include the technique of photon-map rendering where the closest photons to a given one have to be identified and the segmentation phase within a reverse engineering process. We present a new algorithm for k-nearest neighbors computation based on median subdivision and a hashing strategy. The major advantage of our hashing function is that bounds can be established that limit the number of points to be inspected during the search process. Estimates for the asymptotic complexity of our search method are given. Finally we compare our algorithm with a different search strategy based on KD-Trees.

The paper is concerned with the first two steps in a surface reconstruction process. Given a set of 3D points sampled from a physical model the first problem is that of creating a polyhedral approximation of the model. For that the authors introduce an algorithm which extends Boissonnat's (1984) work. It allows the reconstruction of objects with arbitrary genus and proposes an automatic termination procedure. The next step in the process concerns the segmentation of the data points into regions for which each may be fitted by a single surface. They summarize some experiences with a region growing technique based on angle between normals criteria. Using just first order derivative estimations it is shown that the method is able to classify segments into predefined second order surface classes.