Catálogo de publicaciones - libros

Edutainment is a promising approach to motivating students to learn, that is, making learning fun. This paper describes a real world edutainment system that makes learning fun using ubiquitous computing technology. We focus on human-human interaction in the real world as a new approach to edutainment in the next generation. Human actors can teach interactively and flexibly in various ways, by considering players’ (students’) characteristics and the environmental conditions. Involvement of human actors as active characters of a game story in edutainment is an advantage that conventional edutainment does not currently have.

To meet the needs of education in the learning community, an improved adaptive filtering algorithm for teaching resources based on vector space model was proposed in the paper. First, feature selection and pseudo feedback were used to select the initial filtering profiles and thresholds through training algorithm. Then user feedback was utilized to modify the profiles and thresholds adaptively through filtering algorithm. The algorithm had two advantages, the first was that it could carry on self-study to improve the precision; the second was that the execution did not need massive initial texts in the process of filtering. The algorithm was also used in personalized Recommendation service system based on Community E-learning. The result manifested that the algorithm was effective.

This study aims to explore two crucial aspects of collaborative work and learning: the importance of enabling CSCL applications, on the one hand, to capture and structure the information generated by group activity and, on the other hand, to extract the relevant knowledge in order to provide learners and tutors with efficient awareness and support as regards collaboration. To this end, we first identify and define the main types of information generated in on-line group activity and then propose a process for efficiently embedding this information and the knowledge extracted from it into CSCL applications for awareness and feedback purposes. The conceptual model proposed finally gave rise to the design and implementation of a CSCL generic platform, called the Collaborative Learning Purpose Library (CLPL), which serves as a basis for the systematic development of collaborative learning applications and for providing full support to the mentioned process of knowledge management.

This paper presents a novel approach for view-dependent streaming of highly detailed and irregular meshes. In contrast to the con- ventional progressive streaming where the resolution of a model changes in the geometry space, our server firstly maps the 3D surfaces onto the parameter space in which the (GIM) and normal map atlas are obtained by regular re-sampling. By constructing the regular quadtree-based hierarchical representation based on GIM, we accomplish an efficient compression scheme to encode the hierarchical structure that minimize the network communication overhead to facilitate real time mesh update in the large-scale model visualization from clients. Further- more, the encoded nodes can be transmitted in arbitrary order, so the extreme flexibility in transmission could be achieved. In particular, by using normal texture atlas, the rendering result of a partially transmitted model is improved greatly, therefore only the geometry on the silhouette to the current viewpoint requires to be refined and transmitted, which can minimize the amount of data needed to transfer each frame. Experimental results indicate that our approach is efficient enough for a broadcast scenario where one server streams the multiresolution GIM data to multiple clients with different viewpoints.

With the development of network, it’s a common requirement to perceptually access the 3D scenes at a remote server. This demand has not been filled by software standards and their implementations, leaving a wide potential for technical innovations. In the paper, we present a QoS (Quality of Service) controller for remote rendering of 3D contents, aiming at higher real-time performance just as rendering local 3D scenes. The model of our QoS controller is defined by a high-level heuristic function, which has integrated important perceptual factors. And then, guided by the QoS controller, a hybrid transcoding algorithm is developed, which adaptively partitions each 3D model into different modality: the metadata (its compressed base mesh with texture) and a series of refiner meshes. Consequently, the 3D scenes can be adaptively and progressively transmitted to a requesting client, and be fluently displayed at end user’s window in a constant frame rate. Experimental results show the soundness of the proposed controller and algorithm guided by the 3D controller.

In this paper, we present an animation technique based on muscle forces, inverse dynamics and a parameter optimization. We do a crude motion planning in terms of accelerations. By integrating the accelerations given an initial configuration, we obtain all essential kinematic data. We evaluate the quality of motion planning by using various constraints and a performance index evaluated using inverse dynamics. The best motion can be chosen by using a parameter optimization method. The human motion is so complicated that it needs the motions to be coordinated nicely. The planned motion is checked using a criterion which we call the footprint function: ground reaction forces if the body is on the ground, acceleration of the body center as well as energy if it is in the air. In the motion planning for a body in the air, we reduce control variables so that we work with a smaller search space with only feasible motions. Then we include human skeletal and muscle geometry in the footprint function so that we can convert robotic rotary actuators to muscles using static optimization: given a set of joint torques, we distribute them to the eight sets of human low extremity muscles. We search most human-like animated motion from an infinite set of possible motions. We compare these with experimental data. Futhermore, the muscle geometric data obtained from our linear actuator modeling can be used in tissue animation.

Modern graphics application systems have to render many 3D triangle mesh models, thereby facing problems of memory and bandwidth. A general significant solution is to compress the static 3D triangle mesh model at the preprocessing phase, especially the topological information. This paper presents a new triangle meshes compression algorithm, which encodes connectivity information between vertex chains in the meshes that are already stripified. The proposed algorithm provides a new representation that is guaranteed to encode any stripified meshes in less than 1.67 bits/T, and 1.05 bits/T on average. Furthermore, the proposed algorithm employs a short compression/decompression time. Hence it will satisfy real-time rendering constraints.

An algorithm is presented to construct a -continuous B-spline quaternion curve which interpolates a given sequence of unit quaternions on the rotation group (3). We present a method to extend a B-spline interpolation curve to (3). The problem is essentially to find the quaternion control points of the quaternion B-spline interpolation curve. Although the associated constraint equation is non-linear, we can get the accurate quaternion control points according to two additional rules for quaternion computations in . In addition, we provide a point insertion method to construct interpolation curves that have local modification property. The effectiveness of the algorithm is verified by applying it to some examples.

In order to decrease the data volume of 3D textured models in digital museum with significant features preservation, a new simplification method for texture-mesh based on triangle collapse is presented. By introducing the concepts of geometry and texture boundary, the quadric error matrices are improved, and constraint strategies are adopted to retain the feature of geometry boundaries and texture attributes during the simplification process. In addition, a method combining octree-based compression with progressive mesh is given to provide a continuous multi-resolution representation of 3D models. The proposed methods have been successfully applied to progressive transmission and representation system of 3D models in digital museums.

The paper present a novel watermarking algorithm, which is obtained based on a watershed segmentation, one image watermarking algorithm and the addition property of Fourier transform. Firstly, some feature points of 3D meshes are extracted by using an effective watershed segmentation algorithm. Furthermore, in order to keep the mesh transparency, watermark information is preprocessed by using an image watermarking algorithm. Finally, digital watermarking is embedded into those feature points. During extraction, the target model is registered by principal component analysis. We have applied our method to a number of 3D mesh models of different geometric and topological complexity. Those experimental results prove that the algorithm is robust against a variety of attacks, such as rotation, translation, scaling, combinations of the above and re-sort.