Catálogo de publicaciones - libros

Timber-frame building is one of gems of Chinese ancient buildings. Protecting and researching historic buildings by using computer technologies has become a hot topic and urgent task in the field of digital cultural heritage, with increasingly disappearing the timber-frame buildings and their architectural arts and crafts. This paper introduces constructing an information system based on virtual reality technology of timber-frame buildings, and addresses important technologies involved in reconstructing ancient buildings, such as data collection, 3D reconstruction, data organization and management, virtual assembly, and virtual roaming. An integrated solution is proposed and provides a brand-new method for 3D reconstruction, documentation, repair, research and protection of timber-frame buildings.

This paper presents an accelerating hybrid rendering method for complex three-dimensional models using both points and polygons. In circumstance of current PC hardware, our rendering method that integrates advantages of both GBMR (Graphics -Based Modeling and Rendering) and PBMR (Point-Based Modeling and Rendering) works well with complex geometric models which consist of millions of small triangles. In the pre-processing phase, model faces are segmented into regions, triangles and vertex point clouds of each region are stored, and all vertices are sorted in an ascending order according to their importance degree and serialized into a linear structure. In the real-time rendering phase, view-dependent frustum culling and backface clipping are performed; different regions are rendered using triangles or points according to their distance from the viewpoint. These two phases fully utilize the parallelism nature of GPU (Graphic Process Unit) and efficiently implement a continuous multi-resolution rendering for HPPO (Hybrid Point and Polygon Object).

The art of authoring digital multimedia involves collecting and organizing different sorts of media items and transforming them into a coherent presentation. Existing authoring tools for multimedia presentations provide functional support for the authoring process that requires domain knowledge or presentation skills. The authoring process can be enhanced if the authors are supported with decision making, material collection, selection, and presentation composition in generating web presentations. We apply the semantic web technology to generate hypermedia presentations based on media resources retrieved from the web. A discourse model represents the discussion of a subject with a theme supported by a discourse structure that represents the arrangement of contents in a discourse. In this paper we define a discourse model (i.e. Neural Network Architecture) that specifies the knowledge of composing various discourse entities (e.g. Feed-Forward Neural Networks) which enables the building of discourse structures for various themes (e.g. Lecture Notes).

In distributed Geographic Information Systems (GIS) and Web-GIS, remote access to geospatial database is recognized as a time-consuming task. Progressive transmission, a technology dealing with massive amounts of data and slow communication links, can be used to solve this kind of problem because it allows users to start work with a partially delivered dataset. In this paper, we particularly discussed this problem and propose a method, by which we can progressively transmit the sequence of levels of detail of map representation while preserving topological consistency between different levels. We employ the algorithm of minimum number of safe sets to simplify the polyline. In addition, we propose a structure based on Simplicial Multi-Complex and Reactive-tree to avoid data redundancy as well as reduce the space complexity of geospatial database. We show that the volume of coarsest level data can generally be reduced to one tenth of that of the finest level.

Many graphics and visualization applications require fast rendering and animation of fine detailed objects, which are named atomic primitives here. A primitive is a basic graphics shape such as a line or triangle that is directly rasterized into the framebuffer. An atomic primitive is a primitive that is small in size. A large and complex geometric model can be divided into lines, rectangles, and atomic triangles. Here we introduce a new method in the graphics pipeline hardware design to speed up rendering atomic primitives. Specifically, we present a new image-based atomic triangle rasterization algorithm. The algorithm stores the bitmaps of small triangles in memory and quickly retrieves a triangle bitmap when required. Our simulation experiments show that the image-based approach leads to faster rendering in primitive rasterization without sacrificing accuracy. At the same time, clipping, shading, hidden-surface removal, texture mapping, antialiasing, and other basic graphics functions and components are studied with suggested solutions as well. We believe our work is a novel approach that has potential of value and impact in graphics.

In this paper, a 3D mesh retrieval method is proposed based on extracting geometric features of models. The method first finds three principal directions for a model by employing the principal component analysis method, and rotates the model to align it in a reference frame. Then, three sets of planes are used to slice the model along to the directions respectively. Subsequently, three character curves of the model can be obtained and be used as descriptor to key the model in 3D mesh model library. By comparing descriptors of two models, our method can compute similarity of models. Experiences show that our method is rapid, stable and robust to deal with various mesh models with arbitrary geometric and topological complexity.

Human body modeling is essential for virtual environment, sports simulation and digital entertainment. In this paper, we present an improved slice-space deforming algorithm and use it to add shape details extracted from pictures to a reference model in order to get a personalized and structured model. Our method begins with sample image collection and reference image generation followed by silhouette extraction and segmentation. Then it builds a mapping between pixels inside every pair of silhouette segments in the reference image and in the sample image. This novel mapping algorithm is based on a slice space representation that conforms to the intrinsic structure of human body. Finally, based on this mapping, the position of every point in the reference model will be changed. Our deforming algorithm can extract shape details from pictures more precisely and preserve shape connectivity of the reference model in more situations.

Camera setup, calibration and visual based registration of Augmented Reality (AR) based tabletop setups can be a really complicated and time-intensive task. Homography is often used liberally despite its assumption for planar surfaces, where the mapping from the camera to the table can be expressed by a simple projective homography. However, this approach often fails in curved and non-planar surface setups. In this paper, we propose a technique that approximates the values and reduces the tracking error-values by the usage of a neural network function. The final result gives a uniform representation of the camera against combinations of camera parameters that will help in the multi-camera setup. We present the advantages with demonstration applications, where a laser pointer spot and a light from the lamp will be tracked in non planar surface.

In this paper, a novel technique is presented to incorporate vector field based feature extraction schemes into painterly rendering. This approach takes a raster photograph as input and automatically creates a hand-painting style picture. Via techniques formerly used in image segmentation, a vector field representation are generated, identifying color and texture variations at each pixel location, and a series of brush strokes are created with sizes and alignments controlled by the vector field and color matched from the original picture. Moreover, different scale parameters could be utilized to produce several vector fields depicting images features of the original photograph from rough outline to detail. The final output could be rendered first by brushstrokes in the coarsest scale and refined progressively. Unlike conventional techniques that used taking account only of local color gradients, this approach employs multi-scale feature extraction scheme to guide stroke generation with image structure on larger scale.

RTI(Run-Time Infrastructure) provides services for HLA-based distributed simulation, and decides to a great extent the simulation scalability and efficiency. Distributed RTI has good scalability, but its time cost is always higher because of the architecture complexity. BH RTI is based on a distributed RTI architecture. Some techniques are used to overcome the problem of time cost. LoI-based data delivery algorithms are presented to speed up data delivery. PDU(protocol data unit)-function registry and RTI process model are designed to promote the efficiency of data packet processing. Experiment results illustrated that distributed BH RTI depressed the negative influence of architecture complexity and achieved a relatively lower time cost.