WSCG '2010: Poster Paper Proceedings
Permanent URI for this collection
Browse
Recent Submissions
Item Description of image content by means of graph grammars(Václav Skala - UNION Agency, 2010) Zuzaňák, Jiří; Láník, Aleš; Zemčík, Pavel; Skala, VáclavThis paper presents an idea for partial bottom-up parse of image content by use of an attributed graph grammar, in order to achieve effective high-level representation of knowledge contained in image. Terminal nodes of the proposed grammar are formed by image objects (points, lines, and objects detected by classifiers) and areas detected in image by various image processing and segmentation methods. Based on attributes of terminal nodes, each production rule creates derived attributes for high-level representation of lower-level knowledge. Graph that is parsed by graph grammar is constructed in process of knowledge extraction by application of segmentation and image processing algorithms. Created graph is then processed by sequential application of graph grammar rules. Left side of rules is detected by isomorphism detector, and consequent rewrite is performed by rule with highest priority. A part of rewrite process is represented by processing of evaluations of vertices and edges, that describe various properties of objects and their relationships. Further in the paper we present example of attributed graph grammar application in order to describe image content.
Item Procedural modeling in theory and practice(Václav Skala - UNION Agency, 2010) Ullrich, T.; Schinko, C.; Fellner, D. W.; Skala, VáclavProcedural modeling is a technique to describe 3D objects by a constructive, generative description. In order to tap the full potential of this technique the content creator needs to be familiar with two worlds – procedural modeling techniques and computer graphics on the one hand as well as domain-specific expertise and specialized knowledge on the other hand. This article presents a JavaScript-based approach to combine both worlds. It describes a modeling tool for generative modeling whose target audience consists of beginners and intermediate learners of procedural modeling techniques. Our approach will be beneficial in various contexts. JavaScript is a wide-spread, easy-to-use language. With our tool procedural models can be translated from JavaScript to various generative modeling and rendering systems.Item Point cloud lossless compression(Václav Skala - UNION Agency, 2010) Nishio, Koji; Takebayashi, Yusuke; Teshima, Yuji; Kanaya, Takayuki; Kobori, Ken-ichi; Skala, VáclavWe propose a method of downsizing technique for point sampled data which reduces information amount of three dimensional point cloud data. Our method generates clusters and spiral chain lists. Each chain is consists of three dimensional points. After that, we adopt a predictive encoding to compress these chain lists. In addition, we show the effectiveness of our method with same experimental results by comparison with one of conventional methods. From these experiments, our method can reduce the information amount coordinate data to 31.7% of original model.Item Improving the responsiveness in multiplatform collaborative environments(Václav Skala - UNION Agency, 2010) da Silveira, Luiz Gonzaga; Shin-Ting, Wu; Skala, VáclavThis paper addresses the design issues that may improve the responsiveness of a multi-platform collaborative modeling system, for which robustness and awareness are necessary requirements. The key points of our proposal are, whenever possible, (1) to reduce as much as possible the granularity of the transmission data over network; (2) to simplify as much as possible the functional feedbacks for fast screen update; and (3) to avoid as much as possible the network accesses for synchronization. On the basis of these hypotheses, we explore the features of a hybrid groupware architecture and show the feasibility of our proposal. Several latencies are measured to validate our assumptions.Item Efficient acceleration structure layout for 64-bit many-core architectures(Václav Skala - UNION Agency, 2010) Shevtsov, Maxim; Soupikov, Alexei; Skala, VáclavA lot of rendering solutions use an acceleration structure to reduce the complexity of solving geometric proximity search problems. Although acceleration structures are well studied, data exceeding 32 bit address space require an acceleration structure with special properties, such as compact memory layout, efficient traversal capability, memory address space independence, parallel construction capability and 32/64 bit efficiency. We propose a specific memory layout for a kd-tree and methods of processing that data structure handling massive models with the highest efficiency possible. The components of that are easily applied to other hierarchical acceleration structure types as well.Item Post-processing of 3D scanning data for custom footwear manufacture(Václav Skala - UNION Agency, 2010) Savadkoohi, Bita Ture; de Amicis, Raffaele; Skala, VáclavFootwear fitter has been using manual measurement of a long time, but developments of laser scanners in the last few years have now made automatic determination of footwear feasible. Despite the steady increase in accuracy, most available scanning techniques cause some deficiencies in the point cloud and a set of holes in the triangle meshes. Moreover, data resulting from 3D scanning are given in an arbitrary position and orientation in a 3D space. To apply sophisticated modeling operations on these data sets, substantial post-processing is usually required. In this paper, we described an algorithm for filling holes in triangle mesh. First, the advance front mesh technique is used to generate a new triangular mesh to cover the hole. Then, the Poisson equation is applied to optimize the new mesh; then the models are aligned by "Weighted" Principle Component Analysis.Item High resolution 3-D face modeling and model warping(Václav Skala - UNION Agency, 2010) Özüağ, Ersin; Güllü, M. Kemal; Urhan, Oğuzhan; Ertürk, Sarp; Skala, VáclavIn this work, 3-D representations of human faces are obtained at a computer by making use of a DLP projector and a high resolution camera. The system calibration is carried out using a calibration pattern image. A colored structured light approach is adopted in this paper instead of the traditional gray-tone pattern to obtain depth information. A heuristic color correction approach is proposed in this work to improve the performance of the previous approaches. Experiments show that high resolution depth information can be extracted with this approach. Furthermore, a GUI is designed to enable user controlled modifications of the 3-D model.Item Segmentation technique based on object movement for Speech Production simulation(Václav Skala - UNION Agency, 2010) Ohnishi, Youhei; Nozaki, Kazunori; Baba, Ken-ichi; Skala, VáclavIn this paper, we propose the segmentation technique by using Snakes for 4dimentional magnetic resonance imaging (MRI) data of the movements of the oral tract shapes when pronunciation is performed. In the segmentation of the time-sequence MRI images, there is the specific problem that depends on a tongue shape quick deformation. We found that the Optical Flow of image sets is useful criteria for the decision of the geometry of control points in Snakes. Compared to the normal Snakes, our original method which modified Snakes by utilizing Optical Flow demonstrated superior accuracy of the segmentation of 4D MRI data.Item Generation of user interface from characterized code(Václav Skala - UNION Agency, 2010) Kadlec, Jaroslav; Zemčík, Pavel; Skala, VáclavAutomatic generation of the user interface can simplify development of the computer applications. It can help in the development for various target platforms or in simpler testing and algorithm debugging as the user interface can be created for the actual code and platform taking into account many properties. Process of the automatic generation of the user interface can be supported by the data and code characterization. In this paper, an innovative approach using the data and code characterization is presented. The mechanisms and algorithms describing how the data and code characteristics are loaded, the way how objects are transformed into abstract and specific user interface elements, and the process of finalizing user interface is briefly described. As an example, simple media player is described in every step of the user interface generation process.Item Implementing the local binary patterns with SIMD instructions of CPU(Václav Skala - UNION Agency, 2010) Juránek, Roman; Zemčík, Pavel; Herout, Adam; Skala, VáclavUsage of statistical classifiers, namely AdaBoost and its modifications, is very common in object detection and pattern recognition. Performance of such classifiers strongly depends on low level features they use. This paper presents an experimental implementation of the Local Binary Patterns (LBP) that uses SIMD instructions for acceleration. The experiments shows that the proposed implementation is about six times faster than the plain C implementation (i.e. with no special optimizations) and superior to optimized implementations of features with similar descriptive power.Item A bi-quadratic smooth spline surface generation over irregular meshes(Václav Skala - UNION Agency, 2010) Imanparast, Mahdi; Parsaei, Manouchehr; Skala, VáclavA method for generating a smooth spline surface over an irregular mesh is presented. This method generates a smooth spline surface similar to the methods proposed by [Loo94-Pet00, ZZZ+05]. The rules applied to construct the control points in order to achieve the continuity conditions are simple and comprehensible.Item Colorimetric object classification(Václav Skala - UNION Agency, 2010) Wolfram, Hans; Paulus, Dietrich; Skala, VáclavIn order to improve object recognition results, usually several image preprocessings are performed. If color images are used, a color normalization is normally applied. Algorithms for color normalization will be compared to a colorimetric approach found in the literature. Recovering colorimetric values instead of a simple RGB camera output leads to more reliable color images. To this kind of processed object image a basic object recognition approach using different histogram distances is applied. It will be shown that there is an effect on the results of object recognition rates if we use color calibrated images instead of color normalization methods.Item On-line video synchronization based on visual vocabularies(Václav Skala - UNION Agency, 2010) Beran, Vítězslav; Herout, Adam; Zemčík, Pavel; Skala, VáclavThis paper presents the procedure for on-line visual-content-based video synchronization. The motivation of our pioneering work is the existence of several off-line video processing systems employed in video classification or summarization applications, but no evidence of on-line solutions for video analysis. In some applications, the video streams go through the broadcast systems that delay the original video and also distort the original signal. The system that would be able to automatically detect such delays, transmission errors, distortions or broadcast failure is highly required. Our solution employs visual vocabularies that allow signing the video frames by bagof- words. The synchronization procedure is then based on searching for similar frames from different video streams. This paper also overviews the state-of-the-art techniques required for visual vocabulary building and discuss the convenient properties of techniques for real-time on-line systems.Item Interactive streamed media in virtual environment(Václav Skala - UNION Agency, 2010) Běhal, David; Skala, VáclavIn our work, we are dealing with problem of displaying video content under given restrictions. This work is oriented on museums and galleries and their exhibitions. The content has to be secured so it cannot be downloaded. With technology of streaming we offer solution for museums and galleries to display their exhibitions in videos. The transferred data is protected by an encryption in order to reduce the possibility of data leakage. The second challenge is to enrich the video with non-standard interaction elements, to make the exhibition more user-friendly and more interactive. Therefore, we propose using the component called “virtual path” for better orientation in museum. This component also visualizes the process of playing the video. Visitors get the opportunity of moving around the museum in chosen order or skipping corridors, which they do not intent to visit. To do that, we suggest adding some additional information about the scene, which will be displayed on demand.Item A novel content-based image retrieval system based on Bayesian logistic regression(Václav Skala - UNION Agency, 2010) Arias-Nicolás, J. P.; Calle-Alonso, F.; Skala, VáclavIn this work, a novel content-based image retrieval (CBIR) method is presented. It has been implemented and run on “Qatris IManager” [14], a system belonging to SICUBO S.L. (spin-off from University of Extremadura, Spain). The system offers some innovative visual content search tools for image retrieval from databases. It searches, manages and classifies images using four kinds of features: colour, texture, shape and user description. In a typical CBIR system, query results are a set of images sorted by feature similarities with respect to the query. However, images with high feature similarities to the query may be very different from the query in terms of semantics. This discrepancy between low-level features and high-level concepts is known as the semantic gap. The search method presented here, is a novel supervised image retrieval method, based in Bayesian Logistic Regression, which uses the information from the characteristics extracted from the images and from the user’s opinion who sets up the search. The procedure of search and learning is based on a statistical method of aggregation of preferences given by Arias-Nicolás et al. [1] and is useful in problems with both a large number of characteristics and few images. The method could be specially helpful for those professionals who have to make a decision based in images, such as doctors to determine the diagnosis of patients, meteorologists, traffic police to detect license plate, etc.Item Museum guide through annotations using augmented reality(Václav Skala - UNION Agency, 2010) Aracena-Pizarro, Diego; Mamani-Castro, Jovanna; Skala, VáclavThe identification of objects displayed in existing Archaeological Museums, requires the observer to recognize the object or the associated concepts in the thematic environment. Usually, the visitor has a written guide or an expert that accompanies him on his visit. This paper presents an automated guide through annotations in Augmented Reality (AR) unmarked fiducial or free marker, by the recognition of objects shown in the showcase of the museum, showing 2D annotations on the objects. For that purpose, using Head Mounted Displays (HMD) or mobile PDA, the observer can see these annotations. The recognition is based specifically on mathematical relationships that gives us the principal component analysis (PCA), which evolves to create vectors that show more clearly the relationships in the image, which are known as principal component vectors (PCV). These concepts are used to identify archaeological objects and to create a symbolic image database for 3D archaeological objects existing in museum’s showcases, which allows to conform the museum’s guide system in appropriate time. The work helps in shaping an appropriate proposal of recognition to create the symbolic image database of objects, the proper record of annotations on objects, when the mouse click selects it. This record is achieved by robust calculation, and monitoring based on homographies to ensure the insertion of the adequate annotation.Item Augmentation of mean-shift method to facilitate face tracking(Václav Skala - UNION Agency, 2010) Akhriev, Albert; Skala, VáclavIn this paper we introduce an improvement to the well known mean-shift color tracker. On each new frame we estimate the histogram resolution which provides the best separation between color distributions of the object and the background. The optimal resolution is derived from the principle of minimum uncertainty of foreground/background classification. The augmented mean-shift method with variable histogram resolution was applied to the task of face tracking.
