Menu
EN

Vyhledávání

Lidé

doc. Dr. Ing. Radim Šára

Všechny publikace

Online Camera-LiDAR Calibration Monitoring and Rotational Drift Tracking

  • DOI: 10.1109/TRO.2023.3347130
  • Odkaz: https://doi.org/10.1109/TRO.2023.3347130
  • Pracoviště: Vidění pro roboty a autonomní systémy
  • Anotace:
    The relative poses of visual perception sensors distributed over a vehicle's body may vary due to dynamic forces, thermal dilations, or minor accidents. This paper proposes two methods, OCAMO and LTO, that monitor and track the LiDAR-Camera extrinsic calibration parameters online. Calibration monitoring provides a certificate for reference calibration parameters validity. Tracking follows the calibration parameters drift in time. OCAMO is based on an adaptive online stochastic optimization with a memory of past evolution. LTO uses a fixed-grid search for the optimal parameters per frame and without memory. Both methods use low-level point-like features and a robust kernel-based loss function and work with a small memory footprint and computational overhead. Both include a preselection of informative data that limits their divergence. The statistical accuracy of both calibration monitoring methods is over 98%, whereas OCAMO monitoring can detect small decalibrations better, and LTO monitoring reacts faster on abrupt decalibrations. The tracking variants of both methods follow random calibration drift with an accuracy of about 0.03° in the yaw angle.

On Fast Matched Filter for Streak Detection and Ranking

  • Pracoviště: Vidění pro roboty a autonomní systémy
  • Anotace:
    Matched filter is an exceedingly popular method in many fields. In optical astronomy, the common application includes matching streak templates of various lengths and orientations (shape hypothesis). We present a matched filter modification suitable for shorter streaks (less than ca. 100 px) that is faster than current state-of-the-art approaches.

Detection and certification of faint streaks in astronomical images

  • Autoři: Ing. Vojtěch Cvrček, doc. Dr. Ing. Radim Šára,
  • Publikace: VISAPP2019: Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Volume 5. Porto: SciTePress - Science and Technology Publications, 2019. p. 498-509. ISBN 978-989-758-354-4.
  • Rok: 2019
  • DOI: 10.5220/0007399804980509
  • Odkaz: https://doi.org/10.5220/0007399804980509
  • Pracoviště: Vidění pro roboty a autonomní systémy
  • Anotace:
    Fast-moving celestial objects, like near-Earth objects (NEOs), orbiting space debris, or meteors, appear as streaks superimposed over the star background in images taken by an optical telescope at long exposures. As the apparent magnitude of the object increases (the object becomes fainter), its detection becomes progressively harder. We discuss a statistical procedure that makes a binary decision on the presence/absence of a streak in the image which is called streak certification. The certification is based purely on a single input image and a public star catalog, using a minimalistic statistical model. Certification accuracy greater than 90% for streaks of arbitrary orientation, longer than 500 pixels, and the signal-to-background log-ratio is better than −10dB is achieved on the same dataset as in an earlier similar method, whose performance is thus exceeded, especially for close-to-horizontal streaks. We also show that the certification decision indicates detection failure well.

Robust Maximum-likelihood On-line LiDAR-to-Camera Calibration Monitoring and Refinement

  • Autoři: Moravec, J., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings of the 23rd Computer Vision Winter Workshop. Praha: Czech Society for Cybernetics and Informatics, 2018. p. 27-35. ISBN 978-80-270-3395-9.
  • Rok: 2018
  • Pracoviště: Vidění pro roboty a autonomní systémy
  • Anotace:
    In this paper, we present a novel method for online LiDAR–Camera system calibration tracking and refinement. The method is correspondence-free, formulated as a maximum-likelihood learning task. It is based on a consistency of projected LiDAR point cloud corners and optical image edges. The likelihood function is robustified using a model in which the inlier/outlier label for the image edge pixel is marginalized out. The learning is performed by a stochastic on-line algorithm that includes a delayed learning mechanism improving its stability. Ground-truth experimental results are shown on KITTI sequences with known reference calibration. Assuming motion-compensated LiDAR data the method is able to track synthetic rotation calibration drift with about 0.06 degree accuracy in yaw and roll angles and 0.1 degree accuracy in the pitch angle. The basin of attraction of the optimization is about +/- 1.2 degree. The method is able to track rotation calibration parameter drift of 0.02 degree per measurement mini-batch. Full convergence occurs after about 50 mini-batches. We conclude the method is suitable for real-scene driving scenarios.

Faint Streak Detection with Certificate by Adaptive Two-Level Bayesian Inference

  • Pracoviště: Vidění pro roboty a autonomní systémy
  • Anotace:
    It is known that detecting straight streaks from fast moving celestial objects in optical images is an easy problem as long as the streaks are sufficiently long and/or their signal-to-background (SBR) is sufficiently high. At low SBR the situation is different. Since the SBR can be arbitrarily small in practice, a good detection procedure has to provide a detection certificate which is a yes/no answer to the question “does the image contain a streak?” In this paper we pose detection with certificate as a Multi-Level Bayesian Inference (MLBI) problem which is based on Bayesian model selection. We describe the algorithm and show an experimental proof of good behavior on synthetic streaks over real image data. A systematic performance evaluation shows that MLBI confirms and partially exceeds results of state-of-the art methods. In particular, in the class of difficult problem instances with SBR of 0 dB to −5 dB and streak length 10 to 500 pixels, we achieved AUC approximately 0.97, which means that the Bayesian detection certificate is wrong in just 3% of cases.

High accuracy local stereo matching using DoG scale map

  • Autoři: Kitagawa, M., Shimizu, I., doc. Dr. Ing. Radim Šára,
  • Publikace: Machine Vision Applications (MVA), 2017 15th IAPR International Conference on. Piscataway: IEEE, 2017. p. 258-261. ISBN 978-4-901122-16-0.
  • Rok: 2017
  • DOI: 10.23919/MVA.2017.7986850
  • Odkaz: https://doi.org/10.23919/MVA.2017.7986850
  • Pracoviště: Vidění pro roboty a autonomní systémy
  • Anotace:
    Local matching is one of approaches for stereo matching which needs cost aggregation. In Guided Filter based method proposed by Hosni, the cost map is smoothed by Guided Filter using original image as a guiding image. However, the Guided Filter sometimes fails when there are regions whose textures are same but disparities are different. Thus, parameter tuning for filter size of Guided Filter is difficult to obtain the best accuracy. In this paper we propose an algorithm for automatic filter size selection for each pixel of Guided Filter based stereo matching based on the response of the Different of Gaussian (DoG). In our algorithm, we generate the Filter-Size map whose pixel value for each pixel is appropriate filter size. The value of the Filter-Size map is the largest size of the filtering area around the pixel in interest calculated such that more than two edges are not included in filtering area. In our experiments, we evaluated accuracy of Guided Filter based method with our algorithm for selecting filter size compared with the original Guided Filter based method without our algorithm. By using the Middle-bury datasets, the experimental results shows our algorithm's superiority in accuracy.

Stereo Matching for Natural Images with Different Contrasts

  • Autoři: Kitagawa, M., Abe, A., Shimizu, I., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings 1st International Conference on Advanced Imaging. Tokyo: The Imaging Society of Japan, 2015. pp. 654-657.
  • Rok: 2015
  • Pracoviště: Vidění pro roboty a autonomní systémy
  • Anotace:
    In this paper, we propose a robust stereo matching method based on a guided filter using cost function which estimates both similarity of a pixel value and gradient around an image point. We use Moravec's Normalized Cross-Correlation (MNCC) to compute similarity of gradient at corresponding points. We experimentally confirm that the performance of our method for image pairs which have different contrasts is better than in methods based on the guided filter using conventional cost function such as SSD.

Collision Mitigation for Crossing Traffic in Urban Scenarios

  • Autoři: Heck, P., Bellin, J., Matoušek, M., Wonneberger, S., Sychrovský, O., doc. Dr. Ing. Radim Šára, Maurer, M.
  • Publikace: IV 2013: Proceedings of IEEE Intelligent Vehicles Symposium. Piscataway: IEEE, 2013, pp. 559-566. ISSN 1931-0587. ISBN 978-1-4673-2754-1.
  • Rok: 2013
  • DOI: 10.1109/IVS.2013.6629526
  • Odkaz: https://doi.org/10.1109/IVS.2013.6629526
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Current collision mitigation systems focus on rear end collisions. To address the full spectrum of real world accidents, these systems will have to be enhanced to cover more traffic situations. Vehicle to vehicle accidents in crossing traffic situations make up around 25% of accidents in Germany. This paper discusses the requirements and differences compared to rear-end collisions. Presented here is an action concept that takes into account how the impact configuration is changed by breaking the host (impacting) vehicle. Based on this concept the requirements for the detection of crossing traffic were derived. These requirements were met by developing a video system based on a monocular wide field of view camera. It is further shown how this action concept and sensor were integrated into a demonstrator vehicle and evaluated in full scale testing.

FPGA-Accelerated Sliding Window Classifier with Structured Features

  • Autoři: Sychrovský, O., Matoušek, M., doc. Dr. Ing. Radim Šára,
  • Publikace: FPL 2013: Proceedings of the 23rd International Conference on Field Programmable Logic and Applications. Piscataway: IEEE, 2013, pp. 1-4. ISSN 1946-1488. ISBN 978-1-4799-0004-6.
  • Rok: 2013
  • DOI: 10.1109/FPL.2013.6645560
  • Odkaz: https://doi.org/10.1109/FPL.2013.6645560
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Certain classification tasks in computer vision require the classifier response to be computed in every pixel of an image. When combined with large, complex features, it becomes challenging to build such a classifier on a standard PC architecture and achieve real-time performance. We present an FPGA implementation of a car wheel classifier response computation, built as an instantiation of a generic classification system. An interesting optimization problem concerning performance and speed is addressed. Our implementation is running in real-time as a part of a more complex collision mitigation system based on car detection in video data.

RANSACing Optical Image Sequences for GEO and near-GEO Objects

  • Autoři: doc. Dr. Ing. Radim Šára, Matoušek, M., Ing. Vojtěch Franc, Ph.D.,
  • Publikace: Proceedings of the Advanced Maui Optical and Space Surveillance Technologies Conference. Kihei: Maui Economic Development Board, 2013, pp. 924-933. ISSN 2152-4629. Available from: http://www.amostech.com/TechnicalPapers/2013/POSTER/SARA.pdf
  • Rok: 2013
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper describes statistical models and an efficient Monte-Carlo algorithm for detecting tracks of slowly moving objects in optical telescope imagery sequences. The algorithm is based on accurate robust image pre-registration with respect to the star background, hot/warm pixel suppression, extracting dense normalized local image features, pixelwise statistical event detection, segmentation of event maps to putative image primitives, and finding consistent track sequences composed of the image primitives. Good performance at low SNR and robustness of detection with respect to fast or slow-moving thin overhead clouds is achieved by an event detection model which requires collecting at least 10 images of a particular spatial direction. The method does not degrade due to an accumulation of acquisition artifacts if more images are available. The track sequence detection method is similar in spirit to LINE [Yanagisawa et al, T JPN SOC AERONAUT S 2012]. The detection is performed by the RANSAC robust method modified for a concurrent detection of a fixed number of tracks, followed by an acceptance test based on a maximum posterior probability classifier. The statistical model of an image primitive track is based on the consistence between the size and the inclination angle of the image primitive, its image motion velocity, and the sidereal velocity, together with a consistence in relative magnitude. The method does not presume any particular movements of the object, as long as its motion velocity is constant. It can detect tracks without any constraints on their angular direction or length. The detection does not require repeated image transformations (rotations etc.), which makes it computationally efficient. The detection time is linear in the number of input images and, unlike in the LINE proposal method, the number of RANSAC proposals is (theoretically) independent of the number of putative image primitives. The current (unoptimized) experimental implementation run

Spatial Pattern Templates for Recognition of Objects with Regular Structure

  • Autoři: Tyleček, R., doc. Dr. Ing. Radim Šára,
  • Publikace: GCPR 2013: Proceedings of 35th German Conference on Pattern Recognition. Heidelberg: Springer, 2013, pp. 364-374. Lecture Notes in Computer Science. vol. 8142. ISSN 0302-9743. ISBN 978-3-642-40601-0. Available from: http://cmp.felk.cvut.cz/~tylecr1/papers/Tylecek-GCPR2013.pdf
  • Rok: 2013
  • DOI: 10.1007/978-3-642-40602-7_39
  • Odkaz: https://doi.org/10.1007/978-3-642-40602-7_39
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We propose a method for semantic parsing of images with regular structure. The structured objects are modeled in a densely connected CRF. The paper describes how to embody specific spatial relations in a representation called Spatial Pattern Templates(SPT), which allows us to capture regularity constraints of alignment and equal spacing in pairwise and ternary potentials. Assuming the input image is pre-segmented to salient regions the SPT describe which segments could interact in the structured graphical model. The model parameters are learnt to describe the formal language of semantic labelings. Given an input image, a consistent labeling over its segments linked in the CRF is recognized as a word from this language. The CRF framework allows us to apply efficient algorithms for both recognition and learning. We demonstrate the approach on the problem of facade image parsing and show that results comparable with state of the art methods are achieved without introducing additional manually designed detectors for specific terminal objects.

Stochastic Recognition of Regular Structures in Facade Images

  • Autoři: Tyleček, R., doc. Dr. Ing. Radim Šára,
  • Publikace: IPSJ Transactions on Computer Vision and Applications (CVA). 2012, 4 63-70. ISSN 1882-6695.
  • Rok: 2012
  • DOI: 10.2197/ipsjtcva.4.63
  • Odkaz: https://doi.org/10.2197/ipsjtcva.4.63
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We present a method for recognition of structured images and demonstrate it on the detection of windows in facade images. Given an ability to obtain local low-level data evidence on primitive elements of a structure (like window in a facade image), we determine their most probable number, attribute values (location, size) and neighborhood relation. The embedded structure is weakly modeled by pair-wise attribute constraints, which allow structure and attributes to mutually support each other. We use a very general framework of reversible jump MCMC, which allows simple implementation of a specific structure model and plug-in of almost arbitrary element classifiers. We have chosen the domain of window recognition in facade images to demonstrate that the result is an efficient algorithm achieving performance of other strongly informed methods for regular structures.

A Weak Structure Model for Regular Pattern Recognition Applied to Facade Images

  • Autoři: Tyleček, R., doc. Dr. Ing. Radim Šára,
  • Publikace: ACCV 2010: Proceedings of the 10th Asian Conference on Computer Vision, Part I. Berlin: Springer, 2011, pp. 450-463. Lecture Notes in Computer Science. ISSN 0302-9743. ISBN 978-3-642-19314-9.
  • Rok: 2011
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We propose a novel method for recognition of structured images and demonstrate it on detection of windows in facade images. Given an ability to obtain local low-level data evidence on primitive elements of a structure (like window in a facade image), we determine their most probable number, attribute values (location, size) and neighborhood relation. The embedded structure is weakly modeled by pair-wise attribute constraints, which allow structure and attribute constraints to mutually support each other. We use a very general framework of reversible jump MCMC, which allows simple implementation of a specific structure model and plug-in of almost arbitrary element classifiers. The MC controls the classifier by prescribing it 'where to look', without wasting too much time on unpromising locations. We have chosen the domain of window recognition in facade images to demonstrate that the result is an efficient algorithm achieving performance of other strongly informed methods for regular structures like grids, while our general model covers loosely regular configurations as well.

Modeling Symmetries for Stochastic Structural Recognition

  • Autoři: Tyleček, R., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings of the 2nd International Workshop on Stochastic Image Grammars at 2011 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2011, pp. 632-639. ISBN 978-1-4673-0063-6.
  • Rok: 2011
  • DOI: 10.1109/ICCVW.2011.6130302
  • Odkaz: https://doi.org/10.1109/ICCVW.2011.6130302
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We propose a method for stochastic parsing of images with regular structures exhibiting symmetries, such as facades of buildings. The translational symmetry is represented by an array of elements (windows) that is generated with a stochastic grammar which allows structural exceptions and spatial deviations for individual elements. The reflection symmetry of the elements is automatically inferred as a part of the learning process, where a set of random weak features is boosted into a final mixture. A hierarchical probability model is built for the attributed 'words' generated by the proposed grammar. The image parsing result is then found as the most probable interpretation visited with MCMC sampler which is designed to efficiently explore the space of possible configurations.

Real-Time Global Prediction for Temporally Stable Stereo

  • Autoři: Dobiaš, M., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings of the 1st International Workshop on Live Dense Reconstruction from Moving Cameras at 2011 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2011, pp. 704-707. ISBN 978-1-4673-0063-6.
  • Rok: 2011
  • DOI: 10.1109/ICCVW.2011.6130317
  • Odkaz: https://doi.org/10.1109/ICCVW.2011.6130317
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We present a method for calculation of disparity maps from stereo sequences. Disparity map from previous frame is first transferred to the new frame using estimated motion of the calibrated stereo rig. The predicted disparities are validated for the new frame and areas where prediction failed are matched with a traditional stereo matching algorithm. This method produces very fast and temporally stable stereo matching suitable for real-time applications even on non-parallel hardware.

How To Teach Stereoscopic Matching?

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: ELMAR 2010: Proceedings 52th International Symposium ELMAR-2010. Zadar: Croatian Society Electronics in Marine - ELMAR, 2010. pp. 445-453. ISSN 1334-2630. ISBN 978-953-7044-11-4.
  • Rok: 2010
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper describes a simple but non-trivial semi-dense stereoscopic matching algorithm that could be taught in Computer Vision courses. The description is meant to be instructive and accessible to the student. The level of detail is sufficient for a student to understand all aspects of the algorithm design and to make his/her own modifications. The paper includes the core parts of the algorithm in C code. We make the point of explaining the algorithm so that all steps are derived from first principles in a clear and lucid way. A simple method is described that helps encourage the student to benchmark his/her own improvements of the algorithm.

Refinement of Surface Mesh for Accurate Multi-View Reconstruction

  • Autoři: Tyleček, R., doc. Dr. Ing. Radim Šára,
  • Publikace: The International Journal of Virtual Reality. 2010, 9(1), 45-54. ISSN 1081-1451.
  • Rok: 2010
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In this paper we propose a pipeline for accurate 3D reconstruction from multiple images that deals with some of the possible sources of inaccuracy present in the input data. Namely, we address the problem of inaccurate camera calibration by including a method adjusting the camera parameters in a global structure-and-motion problem, which is solved with a depth map for representation that is suitable to large scenes. Secondly, we take the triangular mesh and calibration improved by the global method in the first phase to refine the surface both geometrically and radiometrically. Here we propose surface energy which combines photoconsistency with contour matching and minimize it with a gradient descent method. Our main contribution lies in effective computation of the gradient that naturally regularization and data terms by employing scale space approach. The results are demonstrated on standard high-resolution datasets and a complex outdoor scene.

Depth Map Fusion with Camera Position Refinement

  • Autoři: Tyleček, R., doc. Dr. Ing. Radim Šára,
  • Publikace: CVWW 2009: Computer Vision Winter Workshop 2009. Wien: Pattern Recognition & Image Processing Group, Vienna University of Technology, 2009, pp. 59-66. ISBN 978-3-200-01390-2.
  • Rok: 2009
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We present a novel algorithm for image-based surface reconstruction from a set of calibrated images. The problem is formulated in Bayesian framework, where estimates of depth and visibility in a set of selected cameras are iteratively improved. The core of the algorithm is the minimisation of overall geometric L_2 error between measured 3D points and the depth estimates. In the visibility estimation task, the algorithm aims at outlier detection and noise suppression, as both types of errors are often present in the stereo output. The geometrical formulation allows for simultaneous refinement of the external camera parameters, which is an essential step for obtaining accurate results even when the calibration is not precisely known. We show that the results obtained with our method are comparable to other state-of-the-art techniques.

Languages for Constrained Binary Segmentation Based on Maximum A Posteriori Probability Labeling

  • DOI: 10.1002/ima.20181
  • Odkaz: https://doi.org/10.1002/ima.20181
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    MRF with asymmetric pairwise compatibility constraints between direct pixel neighbors solves a constrained binary image segmentation task. The model is constraining shape and alignment of individual contiguous binary segments by introducing auxiliary labels and their pairwise interactions. Such representation is not necessarily unique. We study several ad-hoc labeling models for binary images consisting of nonoverlapping rectangular contiguous regions. We observed a noticeable increase in performance even in cases when the differences between the models were seemingly insignificant. We use the proposed models for segmentation of windowpanes and windows in orthographically rectified facade images. We show experimentally that even very weak data model in the MAP formulation of the optimal segmentation problem gives very good segmentation results.

Relation between Structural Changes in B-mode Ultrasound Images of Thyroid Parenchyma and the Presence of Thyroid Antibodies in Blood Sample

  • Autoři: Holinka, Ś., doc. Dr. Ing. Radim Šára, Smutek, D.
  • Publikace: Machine Graphics and Vision. 2009, 18(1), 67-82. ISSN 1230-0535.
  • Rok: 2009
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Ultrasonography is a cheap and quick non-invasive medical imaging technique, used as a diagnostic method for autoimmune thyroiditis. This paper shows that B-mode ultrasound images contain weak information related to the presence or absence of antibodies. Ultrasound image analysis is based on textural recognition using probabilistic spatial features. Two studies are performed. The results of the first study show that the spatial texture features we used contain weak information about the presence or absence of TPOAb and TgAb antibodies as measured by conditional entropy. In the second study a classifier is derived from Bayesian decision theory and is tested on a set of 2820 sonograms of 94 subjects. The results of classification to three classes (healthy, thyroiditis with positive antibody test, thyroiditis with negative antibody test) achieved sensitivity 29% and specificity 100% on the test set.

A Maximum Likelihood Surface Normal Estimation Algorithm for Helmholtz Stereopsis

  • Autoři: Guillemaut, J., Mgr. Ondřej Drbohlav, Ph.D., Illingworth, J., doc. Dr. Ing. Radim Šára,
  • Publikace: VISAPP 2008: Proceedings of the Third International Conference on Computer Vision Theory and Applications. Setúbal: INSTICC Press, 2008. pp. 352-359. ISBN 978-989-8111-21-0.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Helmholtz stereopsis is a relatively recent reconstruction technique which is able to reconstruct scenes with arbitrary and unknown surface reflectance properties. Conventional implementations of the method estimate surface normal direction at each surface point via an eigenanalysis, thereby optimising an algebraic distance. We develop a more physically meaningful radiometric distance whose minimisation is shown to yield a Maximum Likelihood surface normal estimate. The proposed method produces more accurate results than algebraic methods on synthetic imagery and yields excellent reconstruction results on real data. Our analysis explains why, for some imaging configurations, a sub-optimal algebraic distance can yield good results.

Fairing of Discrete Surfaces with Boundary That Preserves Size and Qualitative Shape

  • DOI: 10.1007/978-3-540-89639-5_11
  • Odkaz: https://doi.org/10.1007/978-3-540-89639-5_11
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In this paper, we propose a new algorithm for fairing discrete surfaces resulting from stereo-based 3D reconstruction task. Such results are typically too dense, uneven and noisy, which is inconvenient for further processing. Our approach jointly optimises mesh smoothness and regularity. The definition is given on a discrete surface and the solution is found by discrete diffusion of a scalar function. Experiments on synthetic and real data demonstrate that the proposed approach is robust, stable, preserves qualitative shape and is applicable to even moderate-size real surfaces with boundary (0.8M vertices and 1.7M triangles).

Performance Evaluation of Building Detection and Digital Surface Model Extraction Algorithms: Outcomes of the PRRS 2008 Algorithm Performance Contest

  • Autoři: Aksoy, S., Özdemir, B., Eckert, S., Kayitakire, F., Pesaresi, M., Aytekin, O., Borel, C.C., Ing. Jan Čech, Ph.D., Christophe, E., Düzgün, S., Erener, A., Ertugay, K., Hussain, E., Inglada, J., Lefévre, S., Ok, Ö., Koc, D., doc. Dr. Ing. Radim Šára, Shan, J., Soman, J., Ulusoy, I., Witz, R.
  • Publikace: PRRS 2008: Proceedings of the 5th IAPR Workshop on Pattern Recognition in Remote Sensing. Piscataway: IEEE, 2008. p. 37-48. ISBN 978-1-4244-2653-9.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper presents the initial results of the Algorithm Performance Contest that was organized as part of the 5th IAPR Workshop on Pattern Recognition in Remote Sensing (PRRS 2008). The focus of the 2008 contest was automatic building detection and digital surface model (DSM) extraction. A QuickBird data set with manual ground truth was used for building detection evaluation, and a stereo Ikonos data set with a highly accurate reference DSM was used for DSM extraction evaluation. Nine submissions were received for the building detection task, and three submissions were received for the DSM extraction task. We provide an overview of the data sets, the summaries of the methods used for the submissions, the details of the evaluation criteria, and the results of the initial evaluation.

Windowpane Detection based on Maximum Aposteriori Probability Labeling

  • Autoři: Ing. Jan Čech, Ph.D., doc. Dr. Ing. Radim Šára,
  • Publikace: Image Analysis - From Theory to Applications, Proceedings of the 12th International Workshop on Combinatorial Image Analysis (IWCIA'08). Singapore: Research Publishing Services, 2008. pp. 3-11. ISBN 978-3-540-78274-2.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Segmentation of windowpanes in images of building facades is formulated as a task of maximum aposteriori probability labeling. Assuming orthographic rectification of the image, the windowpanes are always axis-parallel rectangles of relatively low variability in appearance. Every image pixel has one of 10 possible labels, and the labels in adjacent pixels are constrained by allowed label configuration, such that the image labels represent a set of non-overlapping rectangles. The task of finding the most probable labeling of a given image leads to NP-hard discrete optimization problem. However, we find an approximate solution using a general solver suitable for such problems and we obtain promising results which we demonstrate on several experiments. Substantial difference between the presented paper and the state-of-the-art papers on segmentation based on Markov Random Fields is that we have a strong structure model, forcing the labels to form rectangles, while other methods do not mode

A Robust Graph-Based Method for The General Correspondence Problem Demonstrated on Image Stitching

  • Autoři: Bujňák, M., doc. Dr. Ing. Radim Šára,
  • Publikace: ICCV 2007: Proceedings of Eleventh IEEE International Conference on Computer Vision. Madison: Omnipress, 2007. ISSN 1550-5499. ISBN 978-1-4244-1631-8.
  • Rok: 2007
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We pose robust matching with parametric and non-parametric constraints as the problem of finding a stable independent set (SIS) in an oriented graph whose vertices are all possible correspondences, whose edges capture the structure of the constraints and whose edge orientation represents pairwise comparison 'is better' based on correspondence quality, including the uncertainty of this comparison. We show SIS possess properties of both robustness and weak optimality. The main contribution of this paper is algorithmic speedup that results from exploiting the dependence between the standard uniqueness constraint and the parametric constraint. The general theory is demonstrated on the example of image stitching using homography model. The algorithm needs at most kN^2 calls of a procedure testing if two ellipse correspondences are consistent with a general homography. The previous known SIS algorithm needed $O(N^4)$ tests. The method gives good results and is fast in practice with k ~ 0.3.

Efficient Sampling of Disparity Space for Fast and Accurate Matching

  • Autoři: Ing. Jan Čech, Ph.D., doc. Dr. Ing. Radim Šára,
  • Publikace: CVPR 2007: Proceedings of the Computer Vision and Pattern Recognition conference. Los Alamitos: IEEE Computer Society, 2007. p. 1-8. ISSN 1053-587X. ISBN 1-4244-1180-7.
  • Rok: 2007
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    A simple stereo matching algorithm is proposed that visits only a small fraction of disparity space in order to find a semi-dense disparity map. It works by growing from a small set of correspondence seeds. Unlike in known seedgrowing algorithms, it guarantees matching accuracy and correctness, even in the presence of repetitive patterns. This success is based on the fact it solves a global optimization task. The algorithm can recover from wrong initial seeds to the extent they can even be random. The quality of correspondence seeds influences computing time, not the quality of the final disparity map. We show that the proposed algorithm achieves similar results as an exhaustive disparity space search but it is two orders of magnitude faster. This is very unlike the existing growing algorithms which are fast but erroneous.

Feasibility Boundary in Dense and Semi-Dense Stereo Matching

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In stereo literature, there is no standard method for evaluating algorithms for semi-dense stereo matching. Moreover, existing evaluations for dense methods require a fixed parameter setting for the tested algorithms. In this paper, we propose a method that overcomes these drawbacks and still is able to compare algorithms based on a simple numerical value, so that reporting results does not take up much space in a paper. We propose evaluation of stereo algorithms based on Receiver Operating Characteristics (ROC) which captures both errors and sparsity. By comparing ROC curves of all tested algorithms we obtain the Feasibility Boundary, the best possible performance achieved by a set of tested stereo algorithms, which allows stereo algorithm users to select the proper method and parameter setting for a required application.

Graph-based Range Image Registration Combining Geometric and Photometric Features

  • Autoři: Shimizu, I., Sugimoto, A., doc. Dr. Ing. Radim Šára,
  • Publikace: SCIA 2007: Proceedings of 15th Scandinavian Conference on Image Analysis. Heidelberg: Springer, 2007. pp. 542-552. ISSN 0302-9743. ISBN 978-3-540-73039-2.
  • Rok: 2007
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We propose a coarse registration method of range images using both geometric and photometric features. The framework of existing methods using multiple features first defines a single similarity distance summing up each feature based evaluations, and then minimizes the distance between range images for registration. In contrast, we formulate registration as a graph-based optimization problem, where we independently evaluate geometric feature and photometric feature and consider only the order of point-to-point matching quality. We then find as large consistent matching as possible in the sense of the matchingquality order. This is solved as one global combinatorial optimization problem. Our method thus does not require any good initial estimation and, at the same time, guarantees that the global solution is achieved.

3D Geometry from Uncalibrated Images

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We present an automatic pipeline for recovering the geometry of a 3D scene from a set of unordered, uncalibrated images. The contributions in the paper are the presentation of the system as a whole, from images to geometry, the estimation of the local scale for various scene components in the orientation-topology module, the procedure for orienting the cloud components, and the method for dealing with points of contact. The methods are aimed to process complex scenes and nonuniformly sampled, noisy data sets.

Can The Presence or Absence of Thyroid Antibodies Be Predicted from Ultrasonographic Images?

  • Autoři: Holinka, Š., doc. Dr. Ing. Radim Šára, Smutek, D.
  • Publikace: Analysis of Biomedical Signals and Images - Proceedings of Biosignal 2006. Brno: VUTIUM Press, 2006, pp. 296-298. ISSN 1211-412X. ISBN 80-214-3152-0.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This study deals with Hashimoto's lymphocytic thyroiditis, namely with the problem of recognition of the presence or absence of thyroid antibodies based on B-mode ultrasonographic images. Leave-one-out results on the training set showed sensitivity of 10% and specificity of 100%. Results on an independent test set of 27 cases showed sensitivity of 29% and specificity of 100%.

FAIR: Towards A New Feature for Affinely-Invariant Recognition

  • Autoři: doc. Dr. Ing. Radim Šára, Matoušek, M.
  • Publikace: ICPR 2006: Proceedings of the 18th International Conference on Pattern Recognition. Los Alamitos: IEEE Computer Society Press, 2006. pp. 412-416. ISBN 0-7695-2521-0.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The first version of FAIR is proposed, which is a low-dimensional image neighborhood descriptor that shows performance comparable to SIFT introduced by Lowe.

Robust Correspondence Recognition for Computer Vision

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: COMPSTAT 2006: Proceedings in Computational Statistics of 17th ERS-IASC Symposium. Heidelberg: Physica-verlag, 2006. pp. 119-131. ISBN 3-7908-1708-2.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We introduce a new robust framework suitable for the task of finding correspondences in computer vision. If the problem domain is general enough, the correspondence problem can seldom employ any well-structured prior knowledge. This leads to tasks that have to find maximum cardinality solutions satisfying some weak optimality condition and a set of constraints. To avoid artifacts, robustness is required to cope with decision under occlusion, uncertainty or insufficiency of data and local violations of prior model. The proposed framework is based on a robust modification of graph-theoretic notion known as digraph kernel.

Ultrasound Image of Chronic Thyroiditis and its Relation to Antithyroid Antibodies

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Holinka, Š., prof. Dr. Ing. Jan Kybic, Tesař, L., Jiskra, J., Maruna, P.
  • Publikace: Proceedings of the 11th Congress of the World Federation for Ultrasound in Medicine and Biology. Amsterdam: Elsevier Science, 2006, pp. 120. ISSN 0301-5629.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We study the relation between ultrasound image of thyroid gland with autoimmune inflammation and levels of antithyroid antibodies.

Complex Correlation Statistic for Dense Stereoscopic Matching

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    A traditional solution of area-based stereo uses some kind of windowed pixel intensity correlation. We introduce a new correlation statistic, which is completely invariant to image sampling, moreover it naturally provides a position of the correlation maximum between pixels. Hereby we can obtain sub-pixel disparity directly from sampling invariant and highly discriminable measurements without any postprocessing of the discrete disparity map.

Globally Convergent Range Image Registration by Graph Kernel Algorithm

  • Autoři: doc. Dr. Ing. Radim Šára, Okatani, I., Sugimoto, A.
  • Publikace: 3DIM 2005: Proceedings of 5th International Conference on 3-D Digital Imaging and Modeling. Los Alamitos: IEEE Computer Society Press, 2005. pp. 377-384. ISBN 0-7695-2327-7.
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Automatic range image registration without any knowledge of the viewpoint requires identification of common regions across different range images and then establishing point correspondences in these regions. We formulate this as a graph-based optimization problem. More specifically, we define a graph in which each vertex represents a putative match of two points, each edge represents binary consistency decision between two matches, and each edge orientation represents match quality from worse to better putative match. Then strict sub-kernel defined in the graph is maximized. The maximum strict sub-kernel algorithm enables us to uniquely determine the largest consistent matching of points. To evaluate the quality of a single match, we employ the histogram of triple products that are generated by all surface normals in a point neighborhood. Our experimental results show the effectiveness of our method for coarse range image registration.

On Reproducibility of Ultrasound Image Classification

  • Autoři: Švec, M., doc. Dr. Ing. Radim Šára, Smutek, D.
  • Publikace: IbPRIA 2005: Proceedings of the Second Iberian Conference on Pattern Recognition and Image Analysis. Berlin: Springer, 2005. pp. 439-446. ISSN 0302-9743.
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Ultrasound B-mode images of thyroid gland were previously analyzed to distinguish normal tissue from inflamed tissue due to Hashimoto's Lymphocytic Thyroiditis. This is a two-class recognition problem. Sensitivity and specificity of 100% was reported using Bayesian classifier with selected texture features. These results were obtained on 99 subjects at a fixed setting of one specific sonograph, for a given manual thyroid gland segmentation and sonographic scan orientation (longitudinal, transversal). To evaluate the reproducibility of the method, sensitivity analysis is the topic of this paper. A general method for determining feature sensitivity to variables influencing the scanning process is proposed. Jensen Shannon distances between modified and unmodified inter- and intra-class feature probability distributions capture the changes induced by the variables. Among selected features, the least sensitive one is found. The proposed sensitivity evaluation method can be used in other pro

Analýza textury ultrasonogramů - nová kvalita v diagnostice tyreopatií?

  • Autoři: Sucharda, P., Smutek, D., doc. Dr. Ing. Radim Šára,
  • Publikace: Časopis lékařů českých. 2004, 143(1), 15-20. ISSN 0008-7335.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Počítačová analýza textury představuje účinný nástroj, schopný bezprostředně po standardním vyšetřeníurčit typ difuzního postižení.

Automatic Disparity Search Range Estimation for Stereo Pairs of Unknown Scenes

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Known disparity search range is crucial for stereo matching tasks, since many algorithms require the disparity search range to be known. Searching over the whole disparity range (i.e. [-image width, image width]) is not only very time consuming (mainly for large images), but even many stereo algorithms do not perform well with unspecified disparity search range. Therefore, automatic estimation of disparity search range for unknown stereo image pairs is highly desired. In low-level image processing (i.e. without knowing any information about the captured scene) this task is very difficult. We propose an approach based on Confidently Stable Matching, which is fast, precise and robust. We demonstrate the algorithm properties on benchmark image sets with known disparity search range as well as on unknown complex scenes.

Citlivostníanalýza přenositelnosti výsledků klasifikace ultrazvukových snímků

  • Autoři: Švec, M., doc. Dr. Ing. Radim Šára,
  • Publikace: Medifórum 3.11.2004, sborník abstraktů. Brno: BVV, 2004, pp. 42. ISBN 80-86607-13-5.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Článek se zabývá klasifikací ultrazvukových snímků. Je navržena metoda ke zjištění přenositelnosti výsledků klasifikace pro změněné podmínky snímání.

Data-Optimal Rectification for Fast and Accurate Stereovision

  • Autoři: Matoušek, M., doc. Dr. Ing. Radim Šára, Hlaváč, V.
  • Publikace: Proceedings of the Third International Conference on Image and Graphics. Los Alamitos: IEEE Computer Society, 2004. pp. 212-215. ISBN 0-7695-2244-0.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In this paper we propose rectification procedure for binocular stereoscopic vision that minimizes the loss of local image neighbourhood discriminability in rectified images. The optimality of the rectification is thus influenced by image contents. Such rectification helps seek for precise dense correspondences.

Helmholtz Stereopsis on rough and strongly textured surfaces

  • Autoři: Guillemaut, J., Mgr. Ondřej Drbohlav, Ph.D., doc. Dr. Ing. Radim Šára, Illingworth, J.
  • Publikace: 3DPVT'04 : Proceedings of the 2nd International Symposium on 3D Data Processing, Visualization, and Transmission. Los Alamitos: IEEE Computer Society Press, 2004. pp. 10-17. ISBN 0-7695-2223-8.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    So far, Helmholtz Stereopsis has been widely applied to objects of smooth geometry and piecewise uniform Bidirectional Reflectance Distribution Function (BRDF). Moreover, for non-convex surfaces the inter-reflection effects have been completely neglected. This paper extends the method to surfaces which exhibit strong texture, nontrivial geometry, and are possibly non-convex.

Joint Non-rigid Motion Estimation and Segmentation

  • Autoři: Flach, B., doc. Dr. Ing. Radim Šára,
  • Publikace: IWCIA '04: Proceedings 10th International Workshop on Combinatorial Image Analysis. Heidelberg: Springer, 2004. pp. 631-638. ISSN 0302-9743.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Usually, object segmentation and motion estimation are considered (and modelled) as different tasks. For motion estimation this leads to problems arising especially at the boundary of an object moving in front of another if e.g. prior assumptions about continuity of the motion field are made. Thus we expect that a good segmentation will improve the motion estimation and vice versa. To demonstrate this, we consider the simple task of joint segmentation and motion estimation of an arbitrary (non-rigid) object moving in front of a still background. We propose a statistical model which represents the moving object as a triangular mesh of pairs of corresponding points and introduce an provably correct iterative scheme, which simultaneously finds the optimal segmentation and corresponding motion field.

Matching Algorithms in Computational Stereoscopic Vision

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Recently it has been recognized that local image modeling in stereo matching brings more discriminable and stable matching features. We introduce a new disparity component matching approach, where the aggregation support regions are defined in disparity space, which guarantees invariance to input image view selection and independence on projective distortions. A rigorous ground-truth evaluation experiment showed that our approach is able to improve overall matching failure rate four-fold while the accuracy is preserved as compared to a standard fixed-size rectangular matching windows approach.

Radiometric Calibration of a Helmholtz Stereo Rig

  • Autoři: Janko, Z., Mgr. Ondřej Drbohlav, Ph.D., doc. Dr. Ing. Radim Šára,
  • Publikace: CVPR 2004: Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE Computer Society Press, 2004. pp. 166-171. ISBN 0-7695-2158-4.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Radiometric calibration of a Helmholtz stereo setup is presented that does not require external light calibration. The problem is shown ill-posed for the case of two cameras but well posed for a greater number of cameras. A calibration procedure that includes a regularizer for the two-camera case is proposed.

Sensitivity Analysis for Reproducibility of Ultrasound Image Classification

  • Autoři: Švec, M., doc. Dr. Ing. Radim Šára, Smutek, D.
  • Publikace: Proceedings of the Computer Vision Winter Workshop 2004 (CVWW'04). Ljubljana: Slovenian Pattern Recognition Society, 2004, pp. 89-98.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Ultrasound B-mode images of thyroid gland were previously analyzed to distinguish normal tissue from inflamed tissue due to Hashimoto's Lymphocytic Thyroiditis. This is a two-class recognition problem. Sensitivity and specificity of 100% was reported using Bayesian classifier with optimal texture features. These results were obtained on 99 subjects at a fixed setting of the sonograph, for a given manual thyroid gland segmentation and sonographic scan type (longitudinal, transversal). To evaluate the reproducibility of the method, sensitivity analysis is the topic of this paper. A general method for determining feature sensitivity to variables influencing the scanning process is proposed. Jensen Shannon distances between modified and unmodified inter- and intra-class feature probability distributions capture the changes induced by the variables. It is shown there are stable features insensitive to small sonograph gain changes and gland segmentation. Features computed from transversal sc

Theory and Robust Algorithm of Trinocular Rectification

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The main contributions are two-fold: Firstly, some theoretical analyses are carried out on trinocular rectification, including the relationship among the three rectified images and their three fundamental matrices, and an geometric interpretation of the 6 free parameters involved in the rectification process. Such results could be used as a theoretical guide to reduce the induced projective distortion. Secondly, under the RANSAC (random sampling consensus) paradigm, a robust trinocular rectification algorithm is proposed. Unlike the traditional ones where only the fundamental matrices are used to rectify images, this algorithm instead uses directly corresponding points for the rectification.

Towards Complete Free-Form Reconstruction of Complex 3D cenes from an Unordered Set of Uncalibrated Images

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper describes a method for accurate dense reconstruction of a complex scene from a small set of high-resolution unorganized still images taken by a hand-held digital camera. A fully automatic data processing pipeline is proposed. Highly discriminative features are first detected in all images. Correspondences are then found in all image pairs by wide-baseline stereo matching and used in a scene structure and camera reconstruction step that can cope with occlusion and outliers. Image pairs suitable for dense matching are automatically selected, rectified and used in dense binocular matching. The dense point cloud obtained as the union of all pairwise reconstructions is fused by local roximation using oriented geometric primitives. For texturing, every primitive is mapped on the image with the best resolution.

Ultrasound of Thyroid Gland - What Is Hidden Inside and Physician Does Not See

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Jiskra, J., Tesař, L.
  • Publikace: Lijecnicki Vjesnik. 2004, 126(2), 57. ISSN 0024-3477.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    It is shown that inflamed and normal thyroid tissue can be differentiated by texture analysis of B-mode sonographic images. Classification success rate of 100 perc. has been achieved in differentiating between an ultrasound image of healthy thyroid gland and an image of gland ith chronic inflammation. The relation between image features and serological markers of thyroid autoimmunity has also been analyzed. A significant correlation between serum levels of thyroid antibodies and texture features was found. These findings suggest the possibility of using sonographic quantitative indicators not only for classification but also as predictors of thyroid antibodies in patient's blood.

A Linear Trinocular Rectification Method for Accurate Stereoscopic Matching

  • Autoři: Zhang, H., Ing. Jan Čech, Ph.D., doc. Dr. Ing. Radim Šára, Wu, F., Hu, Z.
  • Publikace: BMVC 2003: Proceedings of the 14th British Machine Vision Conference. London: British Machine Vision Association, 2003. p. 281-290. ISBN 1-901725-23-5.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In this paper we propose and study a simple trinocular rectification method in which stratification to projective and affine components gives the rectifying homographies in a closed form. The class of trinocular rectifications which has 6 DOF is parametrized by an independent set of parameters with a geometric meaning. This offers the possibility to minimize rectification distortion in a natural way. It is shown experimentally on real data that our algorithm performs the rectification task correctly. As shown on ground-truth data using Confidently Stable Matching, trinocular matching significantly improves disparity map density and mismatch error, both depending on texture strength. Matching results on real complex scenes are reported.

Affine Reconstruction Based on Affinely Corresponding Point Sets

  • Autoři: Zhang, H., Wu, F., doc. Dr. Ing. Radim Šára,
  • Publikace: Computer Vision - CVWW'03 : Proceedings of the 8th Computer Vision Winter Workshop. Prague: Czech Pattern Recognition Society, 2003, pp. 73-76. ISBN 80-238-9967-8.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In this paper we discuss the following observation: Suppose we are given a projective reconstruction Sp of a point set S, in which we can identify subsets Xp subset Sp and Yp subset Sp. Suppose we know that the X and Y must be mutually related by an (unknown) affine mapping Y = Ha X in S. Then it is possible to upgrade Sp to affine reconstruction. We show this problem has up to four solutions. If the internal camera parameters are known to be equal it is possible to choose a single solution. A simple experiment on simulated and real data shows that the problem is numerically feasible.

Computational Stereoscopic Vision

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Recently it has been recognized that local image modeling in stereo matching brings more discriminable and stable matching features. We introduce a new disparity component matching approach, where the aggregation support regions are defined in disparity space, which guarantees invariance to input image view selection and independence on projective distortions. A rigorous ground-truth evaluation experiment showed that our approach is able to increase overall matching failure rate four-fold while the accuracy is preserved as compared to a standard fixed-size rectangular matching windows approach.

Dense Stereomatching Algorithm Performance for View Prediction and Structure Reconstruction

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The knowledge of stereo matching algorithm properties and behaviour under varying conditions is crucial for the selection of a proper method for the desired application. In this paper we study the behaviour of four representative matching algorithms under varying signal-to-noise ratio in six types of error statistics. The errors are focused on basic matching failure mechanisms and their definition observes the principles of independence, symmetry and completeness. A ground truth experiment shows that the best choice for view prediction is the Graph Cuts algorithm and for structure reconstruction it is the Confidently Stable Matching.

Different Types of Image Texture Features in Ultrasound of Patients with Lymphocytic Thyroiditis

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Sucharda, P., Tesař, L.
  • Publikace: Proceedings of the International Symposium on Information and Communication Technologies, ISICT 2003. Dublin: Computer Science Press, 2003, pp. 100-102. ISBN 0-9544145-2-7.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Different Types of Image Texture Features in Ultrasound of Patients with Lymphocytic Thyroiditis

Image Reconstruction Analysis in Lymphocytic Thyroiditis Ultrasound

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Sucharda, P.
  • Publikace: IFMBE Proceedings, World Congress on Medical Physics and Biomedical Engineering. Sydney: University of New South Wales, 2003, ISSN 1727-1983. ISBN 1-877040-14-2.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Image Reconstruction Analysis in Lymphocytic Thyroiditis Ultrasound

Image Texture Analysis of Sonograms in Chronic Inflammations of Thyroid Gland

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Sucharda, P., Tjahjadi, T., Švec, M.
  • Publikace: Ultrasound in Medicine and Biology. 2003, 29(11), 1531-1543. ISSN 0301-5629.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This study shows that inflamed and normal tissues can be differentiated by automatic texture analysis of B-mode sonographic images of thyroid gland.

Relation between Quantitative Description of Ultrasonographic Image and Clinical and Laboratory Findings in Lymphocytic Thyroiditis

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Sucharda, P.
  • Publikace: Endocrine Regulations. 2003, 37(3), 181-187. ISSN 1210-0668.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Relations between measurable properties of B-mode ultrasound images of thyroid gland and clinical and laboratory findings in patients with chronic inflammation of thyroid gland were studied.

Stratified Dense Matching for Stereopsis in Complex Scenes

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Local joint image modeling in stereo matching brings more discriminable and stable matching features. Such features reduce the need for strong prior models (continuity) and thus algorithms that are less prone to false positive artefacts in general complex scenes can be applied. One of the principal quality factors in area-based dense stereo is the matching window shape. As it cannot be selected without having any initial matching hypothesis we propose a stratified matching approach. The window adapts to high-correlation structures in disparity space found in pre-matching which is then followed by final matching. In a rigorous ground-truth experiment we show that Stratified Dense Matching is able to increase matching density 3x, matching accuracy 1.8x, and occlusion boundary detection 2x as compared to a fixed-size rectangular windows algorithm. Performance on real outdoor complex scenes is also evaluated.

Ultrasonogram Analysis of Chronic Autoimmune Thyroiditis Using Non-Heuristic Texture Features

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Sucharda, P.
  • Publikace: Modelling and Control in Biomedical Systems 2003, Proceedings 5th IFAC Symposium. Oxford: Elsevier, 2003. p. 265-269. ISBN 0-08-044159-9.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Ultrasonogram Analysis of Chronic Autoimmune Thyroiditis Using Non-Heuristic Texture Features

A Method for Obtaining 3-Dimensional Facial Expressions and Its Standardization for Use in Neurocognitive studies

  • Autoři: Gur, R.C., doc. Dr. Ing. Radim Šára, Hagendoorn, M., Marom, O., Hughett, P., Macy, L., Turner, T., Bajcsy, R., Posner, A., Gur, R.E.
  • Publikace: Journal of Neuroscience Methods. 2002, 115(2), 137-143. ISSN 0165-0270.
  • Rok: 2002
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    A Method for Obtaining 3-Dimensional Facial Expressions and Its Standardization for Use in Neurocognitive studies

A Photogrammetric Method for Measuring Breathing Movements

  • Autoři: doc. Dr. Ing. Radim Šára, Smutný, V., Veverková, M., Čumpelík, J.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2002. pp. 329-331. ISSN 1211-412X. ISBN 80-214-2120-7.
  • Rok: 2002
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    A Photogrammetric Method for Measuring Breathing Movements

Estimation of the Temporomandibular Joint Position

  • Autoři: Smutný, V., Ing. Jan Čech, Ph.D., doc. Dr. Ing. Radim Šára, Dostálová, T.
  • Publikace: Proceedings of the CVWW'02. Wien: Pattern Recognition & Image Processing Group, Vienna University of Technology, 2002, pp. 306-314.
  • Rok: 2002
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Estimation of the Temporomandibular Joint Position

Estimation of the Temporomandibular Joint Trajectory by Photogrammetry

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Estimation of the Temporomandibular Joint Trajectory by Photogrammetry

Finding the Largest Unambiguous Component of Stereo Matching

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings 7th European Conference on Computer Vision. Berlin: Springer, 2002. pp. 900-914. ISBN 3-540-43744-4.
  • Rok: 2002
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Finding the Largest Unambiguous Component of Stereo Matching

Quantitative Indicators of Sonographic Image of Thyroid Gland and Their Relation to Anti-Thyroid Antibodies in Hashimoto's Lymphocytic Thyroiditis

  • Autoři: Smutek, D., Sucharda, P., doc. Dr. Ing. Radim Šára,
  • Publikace: Health Data in the Information Society. Proceedings of MIE 2002. Amsterdam: IOS Press, 2002. pp. 8-12. ISBN 1-58603-279-8.
  • Rok: 2002
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Sonography is a widely used non-invasive diagnostic tool and its main advantage is low cost in comparison with other diagnostic methods such as immunological analyses. In this work it is presented the relation between a sonographic image of thyroid gland and an immunological status of the patients with Hashimoto's lymphocytic thyroiditis (chronic inflammation of the thyroid gland). The results, evaluated on a set of 740 B-mode sonographic images from 37 subjects, show that raw values of individual image pixels in sonogram of thyroid gland with presence and without presence of anti-thyroid antibodies are significantly different (means 31.87 and 44.56; standard deviations 8.6 and 11.82; t=3.4; p=0.0017) and that they can be used for the prediction of presence of anti-thyreoglobulin and anti-thyreoperoxidasis antibodies.

Specularities Reduce Ambiguity of Uncalibrated Photometric Stereo

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Specularities Reduce Ambiguity of Uncalibrated Photometric Stereo

Stable Matching Based on Disparity Components

Trojrozměrné měření reliéfu dlaždic

  • Autoři: Smutný, V., doc. Dr. Ing. Radim Šára, Krsek, P., Zýka, V.
  • Publikace: Automa. 2002, 8(8-9), 28-29. ISSN 1210-9592.
  • Rok: 2002

Ultrasound Image of the Thyroid Gland and Its Relation to BMI

  • Autoři: Smutek, D., Sucharda, P., doc. Dr. Ing. Radim Šára, Svačina, Š.
  • Publikace: International Journal of Obesity and Related Metabolic Disorders. 2002, 26(Suppl. 1), 202. ISSN 0307-0565.
  • Rok: 2002
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Ultrasound Image of the Thyroid Gland and Its Relation to BMI

Quantitative Tissue Characterization in Sonograms of Thyroid Gland

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Sucharda, P., Svačina, Š.
  • Publikace: Proceedings of the 10th World Congress on Medical Informatics MEDINFO 2001. Amsterdam: IOS Press, 2001. pp. 948. ISBN 1-58603-194-5.
  • Rok: 2001

Stable Monotonic Matching for Stereoscopic Vision

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Robot Vision, Proceedings International Workshop RobVis 2001. Berlin: Springer, 2001. pp. 184-192. ISBN 3-540-41694-3.
  • Rok: 2001

Systematic Construction of Texture Features for Hashimoto's Lymphocytic Thyroiditis Recognition from Sonographic Images

  • Autoři: doc. Dr. Ing. Radim Šára, Smutek, D., Sucharda, P., Svačina, Š.
  • Publikace: Artificial Intelligence in Medicine, Proceedings 8th Conference on Artificial Intelligence in Medicine in Europe, AIME 2001. Berlin: Springer, 2001. pp. 339-348. ISBN 3-540-42294-3.
  • Rok: 2001

Unambiguous Determination of Shape from Photometric Stereo with Unknown Light Sources

Unambiguous Shape from Photometric Stereo with Uncalibrated Light Sources

Accurate Natural Surface Reconstruction from Polynocular Stereo

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings NATO Advanced Research Workshop Confluence of Computer Vision and Computer Graphics. Dordrecht: Kluwer Academic Publishers, 2000, pp. 69-86.
  • Rok: 2000

Chronic Inflammatory Processes in Thyroid Gland: Texture Analysis of Sonographic Images

  • Autoři: Smutek, D., doc. Dr. Ing. Radim Šára, Švec, M., Sucharda, P., Svačina, Š.
  • Publikace: Telematics in Health Care - Medical Infobahn for Europe, Proceedings of the MIE2000/GMDS2000 Congress. Berlin: Quintessenz, 2000, ISSN 1616-2463.
  • Rok: 2000

Diffusion Process Classification in Thyroid Gland Parenchyma Based on Texture Analysis of Sonographic Images: Preliminary Results

  • Autoři: doc. Dr. Ing. Radim Šára, Švec, M., Smutek, D., Sucharda, P., Svačina, Š.
  • Publikace: Proceedings of the Czech Pattern Recognition Workshop. Prague: Czech Pattern Recognition Society, 2000, pp. 45-47. ISBN 80-238-5215-9.
  • Rok: 2000

Integration of Photometric Stereo and Shape from Occluding Contours by Fusing Orientation and Depth Data

  • Autoři: Chen, Y., doc. Dr. Ing. Radim Šára,
  • Publikace: Multi-Image Analysis: Revised Papers, 10th International Workshop on Theoretical Foundations of Computer Vision. Berlin: Springer, 2000. pp. 251-269. ISBN 3-540-42122-X.
  • Rok: 2000

Kvalitativní ukazatele ultrazvukového vyšetření štítné žlázy

  • Autoři: Smutek, D., Tjahjadi, T., doc. Dr. Ing. Radim Šára, Švec, M., Sucharda, P., Svačina, Š.
  • Publikace: Diabetologie, metabolismus, endokrinologie, výživa. 2000, 3(2), 16. ISSN 1211-9326.
  • Rok: 2000

Local Planar Model Verification in a Polynocular Image Set

  • Autoři: Zýka, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings of the Czech Pattern Recognition Workshop. Prague: Czech Pattern Recognition Society, 2000, pp. 107-110. ISBN 80-238-5215-9.
  • Rok: 2000

Physics-Based Method for Determination of Surface Properties from Reflected Light

Polynocular Image Set Consistency for Local Model Verification

  • Autoři: Zýka, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Application of 3D-Imaging and Graph-based Modeling 2000 24-th Workshop of the Austrian Association for Pattern Recognition (ÖAGM/AAPR). Wien: Österreichische Computer Gesselschaft, 2000, pp. 81-88. ISBN 3-85403-142-4.
  • Rok: 2000

Texture Analysis of Sonographic Images for Diffusion Processes Classification in Thyroid Gland Parenchyma

  • Autoři: doc. Dr. Ing. Radim Šára, Švec, M., Smutek, D., Sucharda, P., Svačina, Š.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2000, pp. 210-212. ISBN 80-214-1610-6.
  • Rok: 2000

Fish-Scale Model Verification; A Sub-step of 3D Geometric Model Reconstruction

  • Autoři: Zýka, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings of Workshop 99. Praha: České vysoké učení technické v Praze, 1999, pp. 1.
  • Rok: 1999

Možnosti počítačové analýzy sonogramů difuzních změn ve štítné žláze

  • Autoři: Smutek, D., Sucharda, P., Svačina, Š., doc. Dr. Ing. Radim Šára,
  • Publikace: Diabetologie, metabolismus, endokrinologie, výživa. 1999, 2(1), 35. ISSN 1211-9326.
  • Rok: 1999

Polarization-Based Method for Determination of Surface Properties

Polarization-Based Method for Determination of Surface Properties

Using Polarization to Determine Intrinsic Surface Properties

3-D Data Acquisition and Interpretation for Virtual Reality and Telepresence

  • Autoři: doc. Dr. Ing. Radim Šára, Bajcsy, R., Kamberova, G., Kendall, R.A.
  • Publikace: IEEE and ATR Workshop on Computer Vision for Virtual Reality Based Human Communications. Los Alamitos: IEEE Computer Society Press, 1998. pp. 88-93. ISBN 0-8186-8283-3.
  • Rok: 1998

3D Reconstruction of Environments for Virtual Collaboration

  • Autoři: Bajcsy, R., Enciso, R., Kamberova, G., Nocera, L., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings 4th IEEE Workshop on Applications of Computer Vision. Los Alamitos: IEEE Computer Society Press, 1998. p. 1-10.
  • Rok: 1998

A 3D Geometric Model Acquisition System for a Tele-Collaboration Testbed

  • Autoři: Bajcsy, R., Enciso, R., Kamberova, G., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings ICASE/LaRC/ARO/NSF Workshop on Computational Aerosciences in the 21st Century. Norwell, MA: Kluwer Academic Publishers, 1998, pp. 1-12.
  • Rok: 1998

Fish-Scale Model Refinement for 3D Reconstruction

  • Autoři: Zýka, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Workshop 98. Praha: České vysoké učení technické v Praze, 1998, pp. 167-168.
  • Rok: 1998
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Zpřesnění\ šupinového modelu pro 3D rekonstrukci.

Fish-Scale Surface Model

  • Autoři: doc. Dr. Ing. Radim Šára, Bajcsy, R.
  • Publikace: Proceedings Computer Vision Winter Workshop 1998. Ljubljana: IEEE Slovenia Section, 1998, pp. 1-14. ISBN 961-6062-13-1.
  • Rok: 1998

Polynocular Local Image Dissimilarity for 3D Reconstruction

  • Autoři: Zýka, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings Computer Vision Winter Workshop 1998. Ljubljana: IEEE Slovenia Section, 1998, pp. 15-25. ISBN 961-6062-13-1.
  • Rok: 1998

Three-dimensional reconstruction from a set of video cameras of environments for virtual collaboration

  • Autoři: Bajcsy, R., Kamberova, G., Enciso, R., Nocera, L., Fuchs, H., Welch, G., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings International Symposium on Computer Graphics, Image Processing, and Vision. Los Alamitos: IEEE Computer Society Press, 1998, pp. 36. ISBN 0-8186-9215-4.
  • Rok: 1998

Scale Model Refinement for 3D Reconstruction

  • Autoři: Zýka, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Czech Pattern Recognition Workshop '97. Prague: Czech Pattern Recognition Society, 1997, pp. 121-125.
  • Rok: 1997

Isophotes: the Key to Practable Local Shading

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Computer Analysis of Images and Patterns. Berlin: Springer, 1995. pp. 416-423. ISBN 3-540-60268-2.
  • Rok: 1995

Medicínské aktivity laboratoře počítačového vidění

  • Autoři: Hlaváč, V., Smutný, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Sborník příspěvků na pracovním setkání "Výuka a výzkum v oblasti biomedicínského inženýrství a biotechnologií na FEL ČVUT v Praze". Praha: České vysoké učení technické v Praze, 1995, pp. 73.
  • Rok: 1995

Physically Based Local Shading Analysis

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Workshop 95. Praha: České vysoké učení technické v Praze, 1995, pp. 175-176.
  • Rok: 1995

What Can and What Cannot Be Learned about 3-D Shape from a Single Intensity Image?

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Vision Milestones 95. Proceedings of the Workshop Held at Stift Vorau. Wien: Österreichischen Gesellschaft für Artificial Intelligence, 1995, pp. 101-112. ISBN 3-85027-004-1.
  • Rok: 1995

3D Surface Reconstruction and Its Application in Dentistry

  • Autoři: Hlaváč, V., Dostálová, T., Smutný, V., doc. Dr. Ing. Radim Šára,
  • Publikace: CTU Seminar 94. Praha: České vysoké učení technické v Praze, 1994, pp. 101-102.
  • Rok: 1994

Capturning and Visualisation at the Tooth Occlusion

  • Autoři: Dostálová, T., Hlaváč, V., doc. Dr. Ing. Radim Šára, Smutný, V.
  • Publikace: Proceedings of the 12th international conference Biosignal. Brno: Electrical Engineering Society, 1994, pp. 175-177. ISBN 80-02-00999-1.
  • Rok: 1994

Computer Vision Applications in Dentistry

  • Autoři: Dostálová, T., Hlaváč, V., Pajdla, T., doc. Dr. Ing. Radim Šára, Smutný, V.
  • Publikace: Biomedical Engineering and Biotechnnology. Book of Abstracts. Praha: České vysoké učení technické v Praze, 1994, pp. 11-12.
  • Rok: 1994

Computer Vision Applications in Dentistry - is it Robotics?

  • Autoři: Hlaváč, V., doc. Dr. Ing. Radim Šára, Smutný, V., Dostálová, T.
  • Publikace: Proceedings of the International Symposium "Intelligent Robotics Systems 94". Grenoble: INPG/ENSIEG/LAG, 1994, pp. 246-254.
  • Rok: 1994

Computer Vision Postgraduate Study and Research Development

  • Autoři: Hlaváč, V., Meloun, M., Mirbauer, J., doc. Dr. Ing. Radim Šára,
  • Publikace: CTU Seminar 94. Praha: České vysoké učení technické v Praze, 1994, pp. 99-100.
  • Rok: 1994

Physical Corectness of Local Shading Analysis

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Mustererkennung 1994. Wien: Technische Universität, 1994, pp. 511-518.
  • Rok: 1994

Segmentation of Range Images

  • Autoři: Pajdla, T., Hlaváč, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Acta Stereologica. 1994, 13(2), 459-464. ISSN 0351-580X.
  • Rok: 1994

The Impact of Smooth Surface Perception Capability on the Structure of a Vision System

Three Computer Vision Applications in Dentistry

  • Autoři: Dostálová, T., Hlaváč, V., Pajdla, T., doc. Dr. Ing. Radim Šára, Smutný, V.
  • Publikace: Medical Imaging 1994. Vol. 2. Newport Beach: SPIE, 1994. p. 416-424. ISBN 0-8194-1462-X.
  • Rok: 1994

Digitální rekonstrukce zubní okluze

  • Autoři: doc. Dr. Ing. Radim Šára, Dostálová, T., Hlaváč, V.
  • Publikace: Lékař a technika. 1993, 24(2), 31-34. ISSN 0301-5491.
  • Rok: 1993

Local Shading Analysis Revisited

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings of the 1st Czech Pattern Recognition Workshop. Prague: Czechoslovak Pattern Recognition Society, 1993, pp. 67-74.
  • Rok: 1993

New Image Analysis Algorithms

  • Autoři: Hlaváč, V., Marik, R., Meloun, M., Smutný, V., doc. Dr. Ing. Radim Šára,
  • Publikace: Workshop 93. Praha: České vysoké učení technické v Praze, 1993, pp. 67-68.
  • Rok: 1993

Projection of Smooth Surface Invariants into Intensity Image

  • Autoři: doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings 17th OAGM - Meeting. Graz: Österreichische Computer Gesselschaft, 1993, pp. 83-92.
  • Rok: 1993

Robot Orientation in a 3D World

  • Autoři: Marik, R., Mařík, V., Přeučil, L., doc. Dr. Ing. Radim Šára, Vašek, J.
  • Publikace: Workshop 93. Praha: České vysoké učení technické v Praze, 1993, pp. 79-80.
  • Rok: 1993

Vision-Based Robot Navigation in the 3D World

  • Autoři: Přeučil, L., Vašek, J., Martínek, J., Marik, R., doc. Dr. Ing. Radim Šára,
  • Publikace: Proceedings 4th Conference AI, Simulation and Planning in High Autonomy Systems. Los Alamitos: IEEE Computer Society Press, 1993. p. 200-204. ISBN 0-8186-4020-0.
  • Rok: 1993

Za stránku zodpovídá: Ing. Mgr. Radovan Suk