WSCG 2016: Poster Papers Proceedings
Permanent URI for this collection
Browse
Recent Submissions
Item Distributed video coding for periodic video sequences(Václav Skala - UNION Agency, 2016) Laraba, Moussaab; Benierbah, Said; Khamadja, Mohammed; Skala, VáclavDistributed Video Coding (DVC) is a very active research field that aims to provide simple encoders, needed by many low resources applications. Unfortunately, all the proposed implementations of this type of coding claim that there is no way to design such a system as described by the distributed Source Coding (DSC) theory. With unconvincing results, such a system can only be combined with conventional coding. In this paper, we will show that there is at least one situation where DVC can be applied directly and efficiently. Thus, we adapt and apply the DVC concept to periodic video sequences (PVS). For these sequences, we will propose a new technique to create the side information (SI), where intra coded frames and motion estimation are no longer needed, which makes this technique very simple and yet very effective. The experimental results show a very good performance, and in some cases, we can even outperform H.264 Inter coding.
Item An application of triangle mesh models in detecting patterns of vegetation(Václav Skala - UNION Agency, 2016) Stupariu, Mihai-Sorin; Skala, VáclavThis contribution presents the results of a work in progress, attempting to use discrete curvature for triangle meshes in order to automatically identify specific structures in remote sensing data. Specifically, the focus was on determining isolated trees, on the basis of data acquired through airborne laser scanning. Five methods for discrete curvatures were tested and compared for a triangle mesh derived from a high density point cloud. The best performance was obtained for the mean curvature computed by using the shape operator.Item 2D-clothing shape design using spline curve(Václav Skala - UNION Agency, 2016) Tra Linh, Hoang; Yang, Chuan-Kai; Skala, VáclavFashion design expresses modernness, reflects changes in a society, economy, politics and culture. As a result, fashion also changes very fast and distinctively, and for that reason the improvement and creativity are indispensable. Nowadays, there are numerous fashion design systems/tools. However, they are more about making the clothes software, therefore the provided shape is limited and the way of rendering is stiff since it does not take the shape of the body into account. Tools like Telesia Creator Fashion Design CAD software provides more freedom in the way of representing the clothing shape. However, it adopts a quite unintuitive way to change the shape of a curve. Consequently, we propose the idea of a 2D clothes design system which allows a designer not only to manually create clothes but also to semi-automatically extract a clothes Spline shape from the input image. In other words, our system is able to provide the reference shape for a designer, as well as allows a designer to create his/her work by integrating some provided shape together or even directly modifying the reference shape to create a new one.Item Resilience of luminance based liveness tests under attacks with processed imposter images(Václav Skala - UNION Agency, 2016) Omar, Luma; Ivrissimtzis, Ioannis; Skala, VáclavLiveness tests are techniques employed by face recognition authentication systems, aiming at verifying that a live face rather than a photo is standing in front of the system camera. In this paper, we study the resilience of a standard liveness test under imposter photo attacks, under the additional assumption that the photos used in the attack may have been processed by common image processing operations such as sharpening, smoothing and corruption with salt and pepper noise. The results verify and quantify the claim that this type of liveness tests rely on the imposter photo images being less sharp than live face images.Item Estimating landmarks on 2D images of beetle mandibles(Václav Skala - UNION Agency, 2016) Lê Vãnh, L.; Beurton-Aimar, M.; Salmon, Jean-Pierre; Marie, Alexia; Parisey, N.; Skala, VáclavStudying links between phenotype/genotype and agricultural practices is one of the main topics in agronomy research. Phenotypes can be characterized by informations like age, sex of animals/plants and more and more often with the help of image analysis of their morphology. From now, getting good quality of images for numerous individuals is easy but that leads to design automatic procedures to replace manual exploration of such amount of images. Several bottlenecks have been identified to analyze automatically images. One of them is segmentation of selected area and/or shapes, and another well-known one is setting automatically morphometric landmarks. Landmarks are points on the object which can be used to identify or to classify the objects. It exists a lot of methods to experiment landmarks setting, depending on the image contents. This work has been initiated by using the article of Palaniswamy et al. "Automatic identification of landmarks in digital images"[6]. They proposed a method based on calculus of a probabilistic Hough transform coupling to a template matching algorithm. They applied their method to the Drosophilia wings. In our study, we have gotten a set of 291 beetles . For each one 2D images of 5 different parts of their anatomy have been taken: mandibles left and right, head, pronotum and elytra. The first part of the project was to test how the Palaniswamy’s method could be used to analyze them. We have implemented all the required algorithms to compute positions of mandibles landmarks and compared the obtained results to landmarks which have been manually set by biologists. We will see that even positions automatically obtained are not fully precised, if we used centroid size to characterize mandibles, the size computed from automatic landmarks is closed to this one computed from the manual ones. Future works will focus on definition of a semi-landmarks procedure which would add some features as the measure of the curve between two landmarks.Item Applications of immersive 3D visualization lab(Václav Skala - UNION Agency, 2016) Lebiedź, Jacek; Redlarski, Jerzy; Skala, VáclavThe paper presents Immersive 3D Visualization Lab (at the Faculty of Electronics, Telecommunications and Informatics at Gdańsk University of Technology in Poland) and its applications prepared after its launch in December 2014. The main device of the lab is a virtual reality cubic cave with six screen-walls, where a person can navigate in virtual space using different controllers. The most sophisticated controller is a freely rotating transparent sphere, supported on rollers and equipped with a motion tracking system (something like an omnidirectional hamster wheel). A person's walking motion causes the sphere to revolve and triggers changes in the computer generated 3D images on VR cave screens surrounding the sphere, thus creating an illusion of motion. The projection system is supplemented with a spatial sound generation system.Item Development of the crystal lattice parameter identification method based on the gradient steepest descent method(Václav Skala - UNION Agency, 2016) Shirokanev, Alexandr; Kirsch, Dmitriy; Kupriyanov, Alexandr; Skala, VáclavIn this paper, we propose a new method to solve a problem of crystal lattice parameter identification. The developed method is based on applying the gradient steepest descent method. The two algorithms of crystal lattice parameter identification on the basis of the developed method was proposed. The comparison of the algorithms with the existing method of lattice identification based on estimation of Bravais unit cell parameters is conducted. The experimental results show a significant increase in the crystal lattice parameter identification accuracy for all seven lattice systems.Item Implementation of face recognition methods as a first step for human behaviour analysis in intelligent room(Václav Skala - UNION Agency, 2016) Ronzhin, Alexander; Vatamaniuk, Irina; Železný, Miloš; Skala, VáclavThis paper discuss an intelligent space concepts, goals and developments. The review presents an analysis of seven intelligent meeting rooms, equipment and developed services. One of the main goals of intelligent space is the development of proactive services. Realization of such types of services requires the analysis of current situation in a room as well as behaviour of the audience. Implementation of methods for facial features analysis are the first step for this goal destination. This paper presents an algorithm for facial image normalization and recognition as well as a developed system for automatic registration and identification of meeting participants in the intelligent room. The developed algorithm is intended for processing of low resolution facial images with approximate size of 30x30 pixels. The experiments shows distribution of participants recognition rate by variety of their face sizes. In addition, implementation of blur estimation method at preliminary stage of facial image processing shows an increase of the whole system speed as well as a decrease in face recognition error rate.Item Medical image processing tools for blender with HPC support(Václav Skala - UNION Agency, 2016) Strakoš, Petr; Jaroš, Milan; Karásek, Tomáš; Jarošová, Marta; Krpelík, Daniel; Vašatová, Alena; Burešová, Tereza; Timkovič, Juraj; Štembírek, Jan; Vávra, Petr; Skala, VáclavWe present a medical image processing plug-in in this paper. Our plug-in uses Blender’s environment and adds tools for the medical image processing and 3D model reconstructions and measurements. There are several software solutions to provide these tasks, two of which are used for comparison with our tool in this paper. However, there is not a single application that can perform medical image processing and subsequent 3D model reconstruction and post-processing with possible utilization of High Performance Computing (HPC) resources. In our plug-in we aim to use HPC resources to speed up the most computationally extensive tasks. The plug-in is written in Python and C++. The basic functionality and additional tools are implemented in Phyton. Blender has built-in Python interface, thus the development of the tools is quite fast. The C++ language is used for all time critical functions as it has unrestricted potential in parallel programming that can utilize available HPC resources. In the paper we present two examples, where we demonstrate possibilities of our Blender plug-in. We provide 3D reconstruction of human airway and human orbital floor and compare the results with those provided by 3D Slicer and Osirix.Item Validity of metaphors and views of software visualization for parallel computing(Václav Skala - UNION Agency, 2016) Averbukh, Vladimir L.; Bakhterev, Mikhail O.; Manakov, Dmitriy V.; Skala, VáclavIn this paper approaches to the evaluation of Software Visualization for Parallel Computing are considered on the examples of representation of call graphs and execution traces of parallel programs. The concept of visualization metaphor is described. The visualization metaphors using to depict call graphs and execution traces are surveyed. The validity of visualization techniques is considered on basis of analysis of metaphor properties, Shneiderman’s scheme and other approaches to the evaluation of metaphors and views.Item Reachable sets for dubins car in control problems: physical visualization(Václav Skala - UNION Agency, 2016) Starodubtsev, I. S.; Fedotov, A. A.; Averbukh, V. L.; Patsko, V. S.; Skala, VáclavThe work deals with an application of the 3D-printing to full-size building the reachable sets in control problems. As an example, a simple car model is considered with nonlinear dynamics, three-dimensional phase vector, and scalar control constrained by modulus (Dubins car). Current state of the sytem includes its position in the plane and the velocity heading. The velocity value is given to be constant. The reachable sets are considered “at the terminal instant” and “till the terminal instant”. These sets are nonconvex and their boundaries are not smooth in the whole. Peculiarities of the sets can be better comprehenced being represented in the form of three-dimensional bodies that are built with using the 3D-printing. The sets’ boundaries are presented in the VRML and STL formats. Examples of computation and visualization are given including their 3D-printed copies. Problems are discussed that appear under the 3D-constructing of surfaces with complicated forms.Item Visualizing astrophysics simulations: new results of the COAST project(Václav Skala - UNION Agency, 2016) Thooris, Bruno; Skala, VáclavThe COAST project at CEA/IRFU at Saclay involves astrophysicists and software engineers developing simulation codes in magneto-hydrodynamics and generic tools for data structuration and visualization. Thanks to the new generation of massively parallel mainframes, computing in astrophysics had made huge progress, generating more complexity and much bigger size of numerical simulations. Two software tools have been developed at IRFU for the visualization of massive amounts of data produced by these simulation codes. SDvision is a code deployed in the framework of IDL Object Graphics, to process all results from different simulation codes developed by astrophysicists. PyMSES is a set of Python modules especially optimized for visualizing the complex data produced by our AMR simulation code RAMSES. These codes are suitable for interactive and immersive navigation for the analysis of 3D results and also for videos and stereoscopic movies productions for people at large. In this paper, we present visualizations of recent numerical simulations in astrophysics, more specifically in the domain of Interstellar Medium and Galaxies Formation.Item Image spatial verification using segment intersection of interest points(Václav Skala - UNION Agency, 2016) Bernabeu, Marisa; Pertusa, Antonio; Gallego, Antonio-Javier; Skala, VáclavThis work presents a new spatial verification technique for image similarity search. The proposed algorithm evaluates the geometry of the detected local keypoints by building segments connecting pairs of points and analyzing their intersections in a 2D plane. We show that these intersections remain constant with respect to different geometric transformations (translation, rotation, similarity and affine). Evaluation has been performed obtaining an initial image similarity ranking with a BOF-based methodology, and then using the proposed method for reranking. The presented algorithm (SIIP) has been compared to the RANSAC spatial verification method, showing that it is more efficient and obtains a higher performance on three different datasets.Item Personalized sound zoning for communication means: user studies and evaluation(Václav Skala - UNION Agency, 2016) Fuhrmann, Ferdinand; Amon, Clemens; Leitner, Christina; Maly, Anna; Graf, Franz; Skala, VáclavWe present a new audio user interface for communication means based on automatic sound zoning (ASZ). Highly directive audio reproduction and capturing devices are combined with the output of a newest-generation depth sensor. We particularly use parametric loudspeaker and microphone arrays for the spatial filtering of sound playback and recording. This enables device-less, mobile and ergonomic communication. We evaluated the system using subjective experiments assessing, on the one hand, the immersion properties of the system and, on the other hand, general usability aspects. We could show that users are highly impressed by the capability of such a system and would greatly benefit from the inherent properties.Item Data visualization and evaluation for industry 4.0 using an interactive k-means algorithm(Václav Skala - UNION Agency, 2016) Meiller, Dieter; Niewiera, Florian; Skala, VáclavThe project ISAC@OTH-AW will focus on developing an innovative expert system for data visualization and optimization to produce better manufacturing processes. A mandatory part of the project is the appropriation of industry 4.0 technology benefits, like efficiency, quality and time to market rate, for SMEs (Small and Medium Enterprises).Item Hardware-software embedded face recognition system(Václav Skala - UNION Agency, 2016) Avedillo, M. J.; Barriga, A.; Acasandrei, L.; Calahorro, J. M.; Skala, VáclavThis paper describes the design and implementation of a hardware-software embedded system for face recognition applications in images and/or videos. The system has hardware components to speed up the face detection and recognition stages. It is a system suitable for applications requiring real-time, due that the response times are deterministic and bounded. The system is based on a previous implementation that had accelerated the image capturing process, and the face detection. This paper will focuses in the face recognition acceleration.Item A ballistic missile shutdown point estimation method based on double fitting corrections(Václav Skala - UNION Agency, 2016) Hu, Jing; Gao, Hongtao; Kong, Xiangyu; Dong, Fan; Skala, VáclavFor long-range infrared systems, a new method is proposed in this paper to estimate the shutdown point of ballistic missile. In order to reduce the effect of model error and positioning error of observation point on estimation accuracy, two successive fitting corrections are used in a three-dimensional observation space and a two-dimensional characteristic space respectively. Firstly, the three-dimensional observation data points are fitted by a trajectory plane model, and these points are projected to the trajectory plane for the first correction. Then, a characteristic space l- is set up to describe these projective points in a two-dimensional polar coordinates and the projective points are fitted and corrected by polynomial curves along two axes respectively. Finally, the motion characters of these projective points are converted to the first three-dimensional observation space and the motion state of shutdown point are estimated. The simulation results show that our method is feasible and valid.Item Virtual reality application applied to biomedical models reconstructed from CT scanning(Václav Skala - UNION Agency, 2016) Torner, Jordi; Gómez, Sergi; Alpiste, Francesc; Skala, VáclavVirtual reality-simulated environments have been used for training for more than 40 years. In recent years, an active development of 3D technologies dealing with medical training, planning and guidance has become an increasingly important area of interest in both research and health-care. Moreover, surgery planning and diagnosis using virtual reality (VR) technology has received much attention. User studies show that the method is both effective and easy to use. The purpose of this paper is to present a VR application based on three-dimensional (3-D) surface reconstruction CT imaging for surgical planning and simulation. We propose an interactive tool to improve observation before surgery and increase diagnosis efficiency through virtual reality. This tool allows to obtain a more accurate representation comparing with current technology, such as CT or MRI vision. This cost-effective solution can be easily adopted in any hospital for daily use.Item Free form surface modeling and analysis(Václav Skala - UNION Agency, 2016) Dvořák, Karel; Bílek, Michal; Dvořáková, Stanislava; Skala, VáclavCreating 2D and 3D models in CAD and following creation of drawings are standard procedures for constructing in the current industrial practice. Most models of industrially produced machine parts can be made by basic processes of modeling using construction or hybrid methods. Listed procedures are based on input of a basic – primitive body and its shaping using predefined design elements, combined with the use of construction methods from the default 2D geometry and using methods of extraction, rotation, or dragging a profile along the curve. Requirements of industrial practice and current production technological possibilities often require the implementation part of a more general form, which cannot be created by mentioned methods. In advanced stages of 3D models it is possible to use the methods of free surface shaping, based on creating of parametric, semi parametric or non-parametric network of 3D curves representing the basis for generating surface that defines the boundary of solid body.Item Min-max mipmaps for efficient 2D occlusion culling(Václav Skala - UNION Agency, 2016) Scheckel, Simon; Kolb, Andreas; Skala, Václav3D culling techniques are well established to improve rendering performance, but cannot be applied to 2D games in which the scene is composed of partially transparent textures in a known layer arrangement. Commonly, 2D rendering is achieved in a simple back-to-front blending scheme. This paper discusses options to realize 2D occlusion culling techniques using standard OpenGL functionalities, and introduces a novel 2D culling technique based on min-max mipmaps. We evaluate the performance of the different techniques for different scenarios.
