|
|
Image Restoration Based on Convolutional Neural Networks
Svoboda, Pavel ; Baláž, Teodor (oponent) ; Sojka, Eduard (oponent) ; Zemčík, Pavel (vedoucí práce)
A merit of this thesis is to introduce a unified image restoration approach based on a convolutional neural network which is to some degree degradation type independent. Convolutional neural network models were trained for two different tasks, a motion deblurring of license plate images and a removal of artifacts related to lossy image compression. The capabilities of such models are studied from two main perspectives. Firstly, how well the model can restore an image compared to the state-of-the-art methods. Secondly, what is the model's ability to handle several ranges of the same degradation type. An idea of the unified end-to-end approach is based on a recent development of neural networks and related deep learning in a field of computer vision. The existing hand-engineered methods of image restoration are often highly specialized for a given degradation type and in fact, define state of the art in several image restoration tasks. The end-to-end approach allows to directly train the required model on specifically corrupted images, and, further, to restore various levels of corruption with a single model. For motion deblurring, the end-to-end mapping model derived from models used in computer vision is deployed. Compression artifacts are restored with similar end-to-end based model further enhanced using specialized objective functions together with a network skip architecture. A direct comparison of the convolutional network based models and engineered methods shows that the data-driven approach provides beyond state-of-the-art results with a high ability to generalize over different levels of degradations. Based on the achieved results, this work presents the convolutional neural network based methods suggesting a possibility having the unified approach used for wide range of image restoration tasks.
|
|
|
Scalable Multisensor 3D Reconstruction Framework
Šolony, Marek ; Kneip, Laurent (oponent) ; Sojka, Eduard (oponent) ; Zemčík, Pavel (vedoucí práce)
Realistic 3D models of the environment are beneficial in many fields, from natural or man-made structure inspection, robotic navigation and map building, to the movie industry, in particular, scene survey and special effects integration to scenes. It is common practice to capture the scene with multiple different types of sensors such as monocular, stereoscopic or spherical cameras or 360-degree laser scanners to achieve large coverage of the scene. The advantage of the laser scanners and spherical cameras is that they capture the full surrounding scene as a consistent seamless image. Using easy to operate and manipulate hand-held conventional cameras, the details of the scene obstructed areas are easily covered. The 3D reconstruction consists of three steps--data acquisition, data processing and registration, and refinement of the reconstruction. The contribution of this thesis is a careful analysis of the image registration from several types of cameras~(planar and spherical), as well as 3D laser measurements to obtain an initial estimation of the sensor position and the 3D structure. They are further refined by a unified representation system capable of integrating multisensor measurements and obtain an accurate 3D reconstruction of the environment. The evaluation of the multisensor 3D reconstruction is performed on multiple synthetic, and real-world datasets. The accuracy comparison with commercial multisensor 3D reconstruction software shows that our proposed solution achieves more accurate results. While the commercial solutions are limited to specific type of sensors, our framework can integrate any types of measurements and constraints.
|
|
|
Design and Applications of Special-Purpose Two-Dimensional Visual Markers
Zachariáš, Michal ; Sojka, Eduard (oponent) ; Ftáčnik,, Milan (oponent) ; Herout, Adam (vedoucí práce)
Contemporary visual fiduciary marker systems have a major disadvantage compared to markerless approaches - the camera movement is tightly limited to the space where the markers are. At any frame of the camera image a marker must be large enough to provide sufficient amount of information for detection and identification and at the same time, it must be small enough to fit into the camera's field of view. These requirements are contradictory. This work presents a solution to this problem in a concept of the Marker Field. It is a structure whose presence can be detected in a camera image and the exact location within the field can be recognized based on just any sub-area of defined size. The sub-areas are not disjoint, but they are overlapping to a very large degree, to be identifiable from both close-up and distant views. Different implementations of the marker field concept are explained in this work, together with their intended uses and their advantages and disadvantages. To prove and support the usability of proposed marker fields, this work's second largest part discusses their several real-life applications.
|
|
|
Recognition of Partially Occluded Objects
Krolupper, Filip ; Flusser, Jan (vedoucí práce) ; Sojka, Eduard (oponent) ; Peters, Gabriele (oponent)
In this thesis we focus on partially occluded object recognition under geometric transformations. Objects are represented by their contours. Depending on the kind of geometric transformation and robustness to occlusion we introduce different solutions. Our results are applicable in industry, robotics, 3D vision, forensics, etc. We propose three novel methods for partially occluded object recognition. The major contribution of all our methods is a creation of features. Features are designed to be local and invariant to appropriate geometric transformations. We use mostly standard feature matching to prove properties of designed features. The first method deals only with translation, rotation and scaling (Euclidian transformation) and is based on contour approximation by circle arcs. The parameters of the circle arcs seem to be suitable features. The second method deals with affine transformation and is based on polygonal approximation of contours and, moreover, is robust to additive noise. The second method splits the contour into parts using inflexion points and transforms every part into both normalized shape and position. The parameters of standard shapes of every part are the desired features. The third method deals also with affine transformation. It splits the object into parts using a novel, cutting...
|
|
|
Acceleration of Object Detection Using Classifiers
Juránek, Roman ; Kälviäinen, Heikki (oponent) ; Sojka, Eduard (oponent) ; Zemčík, Pavel (vedoucí práce)
Detection of objects in computer vision is a complex task. One of most popular and well explored approaches is the use of statistical classifiers and scanning windows. In this approach, classifiers learned by AdaBoost algorithm (or some modification) are often used as they achieve low error rates, high detection rates and they are suitable for detection in real-time applications. Object detection run-time which uses such classifiers can be implemented by various methods and properties of underlying architecture can be used for speed-up of the detection. For the purpose of acceleration, graphics hardware, multi-core architectures, SIMD or other means can be used. The detection is often implemented on programmable hardware. The contribution of this thesis is to introduce an optimization technique which enhances object detection performance with respect to an user defined cost function. The optimization balances computations of previously learned classifiers between two or more run-time implementations in order to minimize the cost function. The optimization method is verified on a basic example -- division of a classifier to a pre-processing unit implemented in FPGA, and a post-processing unit in standard PC.
|
|
|
On-line Data Analysis Based on Visual Codebooks
Beran, Vítězslav ; Honec, Jozef (oponent) ; Sojka, Eduard (oponent) ; Zemčík, Pavel (vedoucí práce)
This work introduces the new adaptable method for on-line video searching in real-time based on visual codebook. The new method addresses the high computational efficiency and retrieval performance when used on on-line data. The method originates in procedures utilized by static visual codebook techniques. These standard procedures are modified to be able to adapt to changing data. The procedures, that improve the new method adaptability, are dynamic inverse document frequency, adaptable visual codebook and flowing inverted index. The developed adaptable method was evaluated and the presented results show how the adaptable method outperforms the static approaches when evaluating on the video searching tasks. The new adaptable method is based on introduced flowing window concept that defines the ways of selection of data, both for system adaptation and for processing. Together with the concept, the mathematical background is defined to find the best configuration when applying the concept to some new method. The practical application of the adaptable method is particularly in the video processing systems where significant changes of the data domain, unknown in advance, is expected. The method is applicable in embedded systems monitoring and analyzing the broadcasted TV on-line signals in real-time.
|
|
|
HUMAN ACTION RECOGNITION IN VIDEO
Řezníček, Ivo ; Baláž, Teodor (oponent) ; Sojka, Eduard (oponent) ; Zemčík, Pavel (vedoucí práce)
This thesis focuses on the improvement of human action recognition systems. It reviews the state-of-the-art in the field of action recognition from video. It describes techniques of digital image and video capture, and explains computer representations of image and video. This thesis further describes how local feature vectors and local space-time feature vectors are used, and how captured data is prepared for further analysis, such as classification methods. This is typically done with video segments of arbitrarily varying length. The key contribution of this work explores the hypothesis that the analysis of different types of actions requires different segment lenghts to achieve optimal quality of recognition. An algorithm to find these optimal lengths is proposed, implemented, and tested. Using this algorithm, the hypothesis was experimentally proven. It was also shown that by finding the optimal length, the prediction and classification power of current algorithms is improved upon. Supporting experiments, results, and proposed exploitations of these findings are presented.
|
|
|
Location-aware data transfers scheduling for distributed virtual walkthrough applications.
Přibyl, Jaroslav ; Sochor, Jiří (oponent) ; Sojka, Eduard (oponent) ; Zemčík, Pavel (vedoucí práce)
Data transfers scheduling process is an important part of almost all distributed virtual walkthrough applications. Its main purpose is to preserve data transfer efficiency and rendered image quality. The most limiting factors here are network restrictions. These restrictions can be reduced using multi-resolution data representation, download priority determination and data prefetching algorithms. Advanced priority determination and data prefetching methods use mathematic description of motion to predict next position of a user. These methods can accurately predict only close future positions. In the case of sudden but regular changes in user motion direction (road networks), these algorithms are not sufficient to predict future position with required accuracy and at required distance. In this thesis a systematic solution to data transfers scheduling is proposed which solves also these cases. The proposed solution uses next location prediction methods to compute download priority or additionally prefetch data needed to render a scene in advance. Experiments show that compared to motion functions the proposed scheduling scheme can increase data transfer efficiency and rendered image quality during exploration of tested scene.
|
|
|
Lifting Scheme Cores for Wavelet Transform
Bařina, David ; Kälviäinen, Heikki (oponent) ; Sojka, Eduard (oponent) ; Zemčík, Pavel (vedoucí práce)
The thesis focuses on efficient computation of the two-dimensional discrete wavelet transform. The state-of-the-art methods are extended in several ways to perform the transform in a single loop, possibly in multi-scale fashion, using a compact streaming core. This core can further be appropriately reorganized to target the minimization of certain platform resources. The approach presented here nicely fits into common SIMD extensions, exploits the cache hierarchy of modern general-purpose processors, and is suitable for parallel evaluation. Finally, the approach presented is incorporated into the JPEG 2000 compression chain, in which it has proved to be fundamentally faster than widely used implementations.
|
|
|
Identifikace vozidel na snímcích dopravních situací
Petyovský, Petr ; Sojka, Eduard (oponent) ; Železný, Miloš (oponent) ; Horák, Karel (vedoucí práce)
Cílem této disertační práce je návrh metod pro získání dalších parametrů o vozidle ze snímků reálné dopravní situace ke stávající informaci o RZ vozidla a jeho poloze v měřeném úseku. Úkolem je využít stávající instalace kamerových systémů a na základě dat získaných z těchto zařízení navrhnout nové metody extrakce dalších parametrů o vozidle. Řešení lze rozdělit na dvě skupiny: 1. Metody pro získání příznaků a metody vyhodnocení dat, které povedou k rozpoznání typu vozidla na základě jediného snímku vozidla. 2. Metody pro získání údajů o tvaru vozidla na základě sekvence snímků projíždějícího vozidla
|
host ::
RSS.