Volume 12, number 1-3 (2004)
Permanent URI for this collection
Browse
Recent Submissions
Item Efficient reconstruction of large scattered geometric datasets using the partition of unity and radial basis functions(UNION Agency, 2004) Tobor, Ireneusz; Reuter, Patrick; Schlick, Christopher; Skala, VáclavWe present a new scheme for the reconstruction of large geometric data. It is based on the well-known radial basis function model combined with an adaptive spatial and functional subdivision associated with a family of functions forming a partition of unity. This combination offers robust and efficient solution to a great variety of 2D and 3D reconstruction problems, such as the reconstruction of implicit curves or surfaces with attributes starting from unorganized point sets, image or mesh repairing, shape morphing or shape deformation, etc. After having presented the theoretical background, the paper mainly focuses on implementation details and issues, as well as on applications and experimental results.
Item Interactive solid texturing using point-based multiresolution representations(UNION Agency, 2004) Reuter, Patrick; Schmitt, Benjamin; Schlick, Christopher; Pasko, Alexander; Skala, VáclavThis paper presents an interactive environment for texturing surfaces of arbitrary 3D objects. By uniquely using solid textures and applying them to the surface, we do not require an explicit parameterisation in texture space. Various solid textures can be combined by building a constructive texturing tree of space partitions to define the photometric attributes at each location of the object. Though solid texturing using constructive textures is very powerful, mainly because of its generality, it is quite difficult to use in practice as it is not well suited to interactive tools. To overcome this limitation, we use a multiresolution point-based representation ensuring that texture evaluation and rendering maintains a given frame rate. Our tool is realized as a plugin for the Pointshop3D system. The main advantage of our texturing approach compared to Pointshop3D is that point-based rendering is only used during the interactive texturing step. We always keep a feedback to the initial geometric representation of the object (polygonal mesh, parametric or implicit surface, voxel arrays, or whatever) which means that the final textured object can be easily exported to standard graphics software that cannot directly handle discrete surface points (e.g. CAD systems, photorealistic rendering engines).Item Methods for indexing stripes in uncoded structured light scanning systems(UNION Agency, 2004) Robinson, Alan; Alboul, Lyuba; Rodrigues, Marcos; Skala, VáclavThis paper presents robust methods for determining the order of a sequence of stripes captured in an uncoded structured light scanning system, i.e. where all the stripes are projected with uniform colour, width and spacing. A single bitmap image shows a pattern of vertical stripes from a projected source, which are deformed by the surface of the target object. If a correspondence can be determined between the projected stripes and those captured in the bitmap, a spatial measurement of the surface can be derived using standard rangefinding methods. Previous work has uniquely encoded each stripe, such as by colour or width, in order to avoid ambiguous stripe identification. However, colour coding suffers due to uneven colour reflection, and a variable width code reduces the measured resolution. To avoid these problems, we simplify the projection as a uniform stripe pattern, and devise novel methods for correctly indexing the stripes, including a new common inclination constraint and occlusion classification. We give definitions of patches and the continuity of stripes, and measure the success of these methods. Thus we eliminate the need for coding, and reduce the accuracy required of the projected pattern; and, by dealing with stripe continuity and occlusions in a new manner, provide general methods which have relevance to many structured light problems.Item An approach to improve strip-based multiresolution schemes(UNION Agency, 2004) Ramos, J. F.; Chover, M.; Belmonte, O.; Rebollo, C.; Skala, VáclavTriangle strips have been widely used for static mesh representation because they are optimal for rendering. This primitive reduces the number of vertices sent to the graphics pipeline and the storage costs. We present an approach to improve multiresolution models that takes this connectivity property into account. Our model uses strips both in the data structure and in the rendering stage. It also offers the following features: it is easily implemented, and is efficient and fast.Item A new algorithm for adding color to video or animation clips(UNION Agency, 2004) Pan, Zhigeng; Dong, Zhaohua; Zhang, Mingmin; Skala, VáclavColorizing grayscale video is a useful technique in scientific computing visualization and entertainment. In this paper, we introduce a method to transfer color from a reference image to the whole video. Although the general problem of adding chromatic values to a grayscale image has no exact, objective solution, the current approach attempts to free user from laborious work, except that the user may contribute their skill to reference image. Rather than choosing RGB colors from a palette to color individual components, we resort to anther color space, called lab, which minimizes correlation between channels for many natural scenes. Taking advantage of the correlation between two conjoint frames of video, we track the object and assign the color of it in the preceding frame to that of the posterior one. Experimental results show the algorithm works quite well.Item A hybrid approach to rendering handwritten characters(UNION Agency, 2004) Su, Sara L.; Wu, Chenyu; Xu, Ying-Qing; Shum, Heung-Yeung; Skala, VáclavWith the growing popularity of pen-based computers comes the need to display clear handwritten characters at small sizes on low-resolution displays. This paper describes a method for automatically constructing hinted TrueType fonts from on-line handwriting data. Hints add extra information to glyph outlines in the form of imperative constraints overriding side effects of the rasterization process. We use an aggressive matching strategy to find correspondences between an input glyph and a previously-hinted template, considering both global and local features to allow hinting even when they differ in shape and topology. Recognizing that stroke width statistics are among features that characterize a person’s handwriting, we recalculate global values in the control value table (CVT) before transfer to preserve the characteristics of the original handwriting.Item 3D pose estimation using coupled snakes(UNION Agency, 2004) Hohnhäuser, Benjamin; Hommel, Günter; Skala, VáclavAn extension to the classical concept of active contour models is proposed. Besides the introduction of a new technique to couple several contours and treat them as one, the coupling of active contours beyond image domains is presented. The coupling is realized by a force, that controls the mutual attraction or repulsion of each active contour, depending on its definition. The coupling across the borders of images acquired by different sources provides a higher integration of information for each extracted image feature implying an appropriate registration of the images.Item Registration tasks for a hybrid tracking system for medical augmented reality(UNION Agency, 2004) Schwald, Bernd; Seibert, Helmut; Skala, VáclavTracking technologies for medical Virtual or Augmented Reality (VR/AR) applications have to fulfill demanding requirements. A hybrid tracking system can merge positive attributes of different tracking technologies, for example optical and electromagnetic, and try to compensate the bad ones. The alignment of two or more tracking systems by means of transforming all measured data from different systems into one common coordinate system is a fundamental precondition. Usage and processing of time stamps assigned to each measurement is a very effective element in order to achieve a higher accuracy in the alignment procedure and during later use. The algorithms for the alignment of tracking systems can also be used for the registration of the patient, another typical task in medical VR/AR applications. This paper describes both procedures in detail, including algorithms and results from test sequences. The representative tests illustrate, that the proposed algorithms provide satisfying results. An overview of the integration of the hybrid tracking system in a medical AR system is given. Adverse conditions like environmental influences on the tracking systems and the identification of corresponding measurements of two tracking systems are discussed.Item Hand gesture recognition for human-machine interaction(UNION Agency, 2004) Sánchez-Nielsen, Elena; Antón-Canalís, Luis; Hernández-Tejera, Mario; Skala, VáclavEven after more than two decades of input devices development, many people still find the interaction with computers an uncomfortable experience. Efforts should be made to adapt computers to our natural means of communication: speech and body language. The PUI paradigm has emerged as a post-WIMP interface paradigm in order to cover these preferences. The aim of this paper is the proposal of a real time vision system for its application within visual interaction environments through hand gesture recognition, using general-purpose hardware and low cost sensors, like a simple personal computer and an USB web cam, so any user could make use of it in his office or home. The basis of our approach is a fast segmentation process to obtain the moving hand from the whole image, which is able to deal with a large number of hand shapes against different backgrounds and lighting conditions, and a recognition process that identifies the hand posture from the temporal sequence of segmented hands. The most important part of the recognition process is a robust shape comparison carried out through a Hausdorff distance approach, which operates on edge maps. The use of a visual memory allows the system to handle variations within a gesture and speed up the recognition process through the storage of different variables related to each gesture. This paper includes experimental evaluations of the recognition process of 26 hand postures and it discusses the results. Experiments show that the system can achieve a 90% recognition average rate and is suitable for real-time applications.Item Palmprint verification with moments(UNION Agency, 2004) Pang, Y. H.; Andrew, T. B. J.; David, N. C. L.; Hiew, F. S.; Skala, VáclavPalmprint verification is an approach for verifying a palmprint input by matching the input to the claimed identity template stored in a database. If the dissimilarity measure between the input and the claimed template is below the predefined threshold value, the palmprint input is verified possessing same identity as the claimed identity template. This paper introduces an experimental evaluation of the effectiveness of utilizing three well known orthogonal moments, namely Zernike moments, pseudo Zernike moments and Legendre moments, in the application of palmprint verification. Moments are the most commonly used technique in character feature extraction. The idea of implementing orthogonal moments as palmprint feature extractors is prompted by the fact that principal features of both character and palmprint are based on line structure. These orthogonal moments are able to define statistical and geometrical features containing line structure information about palmprint. An experimental study about verification rate of the palmprint authentication system using these three orthogonal moments as feature descriptors has been discussed here. Experimental results show that the performance of the system is dependent on the moment order as well as the type of moments. The orthogonal property of these moments is able to characterize independent features of the palmprint image and thus have minimum information redundancy in a moment set. Pseudo Zernike moments of order of 15 has the best performance among all the moments. Its verification rate is 95.75%, which also represents the overall performance of this palmprint verification system.Item Robust tracking of athletes using multiple features of multiple views(UNION Agency, 2004) Misu, Toshihiko; Gohshi, Seiichi; Izumi, Yoshinori; Fujita, Yoshihiro; Naemura, Masahide; Skala, VáclavThis paper presents a robust and reconfigurable object tracker that integrates multiple visual features from multiple views. The tandem modular architecture stepwise refines the estimate of trajectories of the objects in the world coordinates using many plug-ins that observe various features such as texture, color, region and motion in 2D images acquired by the cameras. One of the most important features of our proposed method is that each plug-in innovates the trajectories not only by back-projecting 2D observations of the features, but also by weighting them adaptively to their self-evaluated reliability. In the paper, the architecture of the system and that of the plug-ins are formulated. The behavior and robustness against occlusion are also shown through experiments with football-game sequences.Item Computer-assisted analysis of in-vitro vasculogenesis and angiogenesis processes(UNION Agency, 2004) Lamberti, Fabrizio; Gamba, Andrea; Montrucchio, Bartolomeo; Skala, VáclavIn recent years many experimental investigations have been carried out on vasculogenesis and angiogenesis, the mechanisms of blood vessels formation. Neovascularization is a hallmark of embryogenesis and many other physiological processes, such as wound healing or endometrium vascularization during the mestrual cycle. Angiogenesis also plays a key role in tumor growth, tumor metastasis and other pathological processes including many inflammatory diseases. Understanding biological phenomena regulating angiogenesis is therefore essential for clinical treatment of cancer and other angiogenesis-dependent diseases. This paper describes the development of a fully-automated computerized angiogenesis analysis system, which enables angiogenesis parameters to be quantified in an in-vitro model. The proposed methodology works on phase contrast microscopy snap photographs of cultured endothelial cell plates, and extracts a detailed graph-based representation of the blood vessels network thus supporting accurate angiogenesis parameters measurement.Item Feature-point based matching: a sequential approach based on relaxation labeling and relative orientation(UNION Agency, 2004) Galo, Maurizio; Tozzi, Clésio L.; Skala, VáclavThis paper presents a solution for the problem of correspondence and relative orientation (RO) estimation for a pair of images. The solution is obtained by relaxation labeling using multiple metrics applied to image primitives. Besides the use of metrics based on radiometric elements (intensities, gradient, and coefficient of correlation), and on geometry (distance ratios), two additional metrics are considered. One is based on angular relations between primitives and another based on the volume of Matching Parallelepiped (MP) that allows the inclusion of the epipolar geometry constraint directly in the similarity and compatibility computation. The proposed solution was applied to synthetic and real images. The results showed that the use of multiple metrics contribute to the automation of the correspondence process and RO determination, even considering pairs of images subject to convergence, rotation, differences in scale, and presence of repetitive patterns.Item Efficient generation of triangle strips from triangulated meshes(UNION Agency, 2004) van Kaick, Oliver Matias; da Silva, Murilo Vicente Gonçalves; Pedrini, Hélio; Skala, VáclavThis paper presents a fast algorithm for generating triangle strips from triangulated meshes, providing a compact representation suitable for transmission and rendering of the models. A data structure that allows efficient triangle strip generation is also described. The method is based on simple heuristics, significantly reducing the number of vertices used to describe the triangulated models. We demonstrate the effectiveness and speed of our method comparing it against the best available program.Item Image registration exploiting five-point coplanar perspective invariant and maximum-curvature point(UNION Agency, 2004) Putjarupong, P.; Pintavirooj, C.; Withayachumnankul, W.; Sangworasil, M.; Skala, VáclavImage registrations have been a subject of extensive study over the last decade. They appear in numerous applications including Computer Vision, Pattern Recognition, Medical Image Analysis and Remotely Sensed Image Processing. We introduce a non-iterative geometric-based method to register an image using a novel set of geometric landmarks residing on an extracted 2D contours from the image. These landmarks are intrinsic and are computed from the differential geometry of the curve. We exploit the invariant properties of maximumcurvature points that are local and preserved under the affine and some perspective transformation. Geometric invariant exploits coplanar five-point invariant constructed from 4 consecutive landmarks and a centroid. This invariant is preserved not only in affine map but weak perspective map as well. To reduce the sensitivity of the landmarks to noise, we use a B-Spline surface representation that smoothes out the curve prior to the computation of the landmarks. The alignment is achieved by establishing correspondences between the landmarks after a conformal sorting based on derived absolute invariant. The experiments have shown that the purposed methods are robust and promising even in the presence of noise.Item Efficient collision detection between 2D polygons(UNION Agency, 2004) Jiménez, Juan José; Segura, Rafael J.; Feito, Francisco R.; Skala, VáclavCollision detection between moving objects is an open question which raises major problems concerning its algorithmic complexity. In this paper we present a polygon collision detection algorithm which uses polygon decomposition through triangle coverings and polygon influence areas (implemented by signs of barycentric coordinates). By using influence areas and the temporal and spatial coherence property, the amount of time needed to detect a collision between objects is reduced. By means of these techniques, a valid representation for any kind of polygon is obtained, whether concave or convex, manifold or non-manifold, with or without holes, as well as a collision detection algorithm for this type of figures. This detection algorithm has been compared with the well-known PIVOT2D [Hof01] one and better results have been achieved in most situations. This improvement together with its possible extension to 3D makes it an attractive method because pre-processing of the polygons is no longer necessary. Besides, since this method uses sign operations, it proves to be a simple, more efficient and robust method.Item Efficient trimmed NURBS tessellation(UNION Agency, 2004) Balázs, Ákos; Guthe, Michael; Klein, Reinhard; Skala, VáclavInteractive rendering of trimmed NURBS models is of great importance for CAD systems. For this the model needs to be transformed into a polygonal representation. This transformation can be either performed in a preprocessing step, at the cost of losing the capability to edit the surfaces, or on the fly during rendering. Since the number of frames per second is usually critical, efficient on the fly tessellation of trimmed NURBS surfaces is very important for interactive rendering and editing of such models. In this paper we present an efficient – with respect to both runtime and to the number of generated triangles – tessellation algorithm for trimmed NURBS surfaces that is capable of guaranteeing a specified geometric approximation error. When affordable by the subsequent steps in the pipeline, an approximate error of the tessellation can also be used leading to fewer triangles.Item Color correction methods with applications to digital projection environments(UNION Agency, 2004) Yin, Jianfeng; Cooperstock, Jeremy R.; Skala, VáclavDue to differing optics, sensor characteristics, and hardware processing employed by video cameras, the resulting colors produced by two cameras can be very different, thus complicating the task of computer vision applications. While various color correction methods exist to deal with this problem, most involve strong assumptions, such as constant illumination, that are, in general, unsatisfied in complex environments. To address the problem of color correction in a less restrictive manner, we propose the use of neural networks, which can easily be trained and which produce excellent results. We compare these results with other methods.Item Blending textured images using a non-parametric multiscale MRF method(UNION Agency, 2004) Tiddeman, Bernard; Skala, VáclavIn this paper we describe a new method for improving the representation of textures in blends of multiple images based on a Markov Random Field (MRF) algorithm. We show that direct application of an MRF texture synthesis algorithm across a set of images is unable to capture both the "averageness" of the global image appearance as well as specific textural components. To overcome this problem we vary the width of the Parzen window (used to smooth the conditional probability distribution of the pixel's intensity) as a function of scale, thus making lower pyramid resolutions closer to the Gaussian mean, while maintaining the high resolution textures. We also show that approximating the maxima of the conditional probability distributions with a weighted-average produces very similar results with a significant increase in speed.
