|
|
Optical Field and Hologram Synthesis
Šulgan, Marián ; Seeman, Michal (referee) ; Zemčík, Pavel (advisor)
The main focus of this work is to implement appropriate chosen method of optical field synthesis in several forms, compare their results and confirm its validity by reconstruction of hologram generated from calculated optical field. For its complexity and demanding calculations, various accelerations are employed - linear approximation, parallelization, using technology C++ AMP, which facilitates opportunity to use GPU for massive parallelism. The work begins with explanation of basic holography principles followed by design of application offering solutions to set problems. The final part explains the implementation of application programmed in C++, testing process, gained results and their evaluation. The text ends with final conclusions regarding completion of set tasks and possible future expanding or improving.
|
|
|
Real-Time Image Processing
Polok, Lukáš ; Sumec, Stanislav (referee) ; Zemčík, Pavel (advisor)
This work is concerned with acceleration of two-dimensional image processing using programmable graphical hardware (or GPU's). Set of implemented operations includes some basic, mostly atomic, image processing functions. However, library is designed in such a manner so it can be easily extended by adding new functions. Implementation of graphical algorithms using GPU's profits from their high computational power and fast development cycles. For situations where target hardware doesn't support programmable shading, required by the library, reference "C" language implementation is available.
|
|
|
Ray Tracing on CUDA Architecture
Bidmon, Lukáš ; Polok, Lukáš (referee) ; Bařina, David (advisor)
This work presents utilization of CUDA capable graphic cards for ray tracing. First, the classic recursive ray tracing algorithm is presented and necessary math is explained for implemented objects. nVidia CUDA architecture is introduced in next chapter with explained differences from CPU computations. Following is the implementation scheme where modifications necessary for CUDA are discussed. Implementation chapter covers details about flow of the program and memory usage. Finally the CPU and GPU testing results are presented.
|
|
|
Object Tracking in Video Using CUDA
Schery, Miroslav ; Polok, Lukáš (referee) ; Herout, Adam (advisor)
This bachelors thesis is focused on implementation of particle filter algorithm in CUDA technology for the purpose of computation acceleration. It describes object tracking methods in video and is especialy aimed at particle filter. It also describes the CUDA architecture. Implementation and optimalization techniques used in application are explained. The work ends with performing speed tests and the tracking ability of the algorithm is verified on various videos.
|
|
|
Radiosity on GPU
Šabata, David ; Nečas, Ondřej (referee) ; Polok, Lukáš (advisor)
This work deals with the radiosity algorith, a global illumination method and its implementation in OpenGL and OpenCL libraries. At rst two of the most common global illumination methods will be presented, considering radiosity as the main topic, including its usage in realtime rendering. An introduction to the libraries used will be next, followed by description of the application implemented in C++ language. In the end the ndings of this thesis and its possible improvements will be discussed.
|
|
|
Particle Systems
Bedecs, Jakub ; Maršík, Lukáš (referee) ; Kajan, Rudolf (advisor)
This bachelor's thesis deals with the implementation of particle systems with the usage of calculation power of GPU. The purpose of this work is to describe all important facts about the particle systems construction and to show up various possibilities of its usage. It analysis the abilities of modern shaders and their usage for calculation of particles movement. The basis of this work is the analysis of the implemented application, which is able to dynamically change all parameters of the system.
|
|
|
GPU Image Processing Library
Čermák, Michal ; Španěl, Michal (referee) ; Smrž, Pavel (advisor)
This work is concerned with architecture of recent Nvidia graphics cards and application programming interface CUDA. That is used to create accelerated image processing library. It place emphasis on testing performance gain compassion with high optimized and used OpenCv library.
|
|
|
MRI Data Processing Acceleration on GPU
Kešner, Filip ; Nečas, Ondřej (referee) ; Polok, Lukáš (advisor)
This BSc Thesis was performed during a study stay at the Universita della Svizzera italiana, Swiss. The identification of trajectories of neuron fibres within the human brain is of great importance in many medical applications as the neural diagnostics, neuronavigation, treatment of epilepsy, surgical removal of tumors and etc. By using diffusion MRI-data as input, and by employing Monte-Carlo like methods, possible trajectories are generated and the most likely ones can be visualized. These can serve as input for advanced medical diagnosis and treatments. Due to the huge amount of data to be analyzed and many iterations, this is a time consuming process. For the purposes such as statistical analysis and comparsion over several datasets or several patients, computational time requirements are enourmous. Faster diagnosis can improve routine throughput and provide earlier treatment of illness. At this time, there exists only a very few implementations of neural tractography sof tware. For probabilistic neural tractography is the list of software even thiner. Today's implementations using standard serial CPU execution suffer from high time consumption. The goal is to provide an efficient implementation which makes use of GPGPUs and exploits parallelism in the method. For the GPU implementation, a comparsion of CUDA and OpenCL technologies will be provided, using the more suitable one.
|
|
|
Digital Filter Design on GPU
Vaverka, Filip ; Maršík, Lukáš (referee) ; Polok, Lukáš (advisor)
This thesis shows one of the approaches to the design of ditigal filters with infinite impulse response and specified order. The proposed solution is based on an evolutionary genetic algorithm and therefore allows for direct filter design from its specification. Its main contribution to this subject is that the implementation is parallel and it is acceleraded by GPU. The filters are designed in cascade representation. It also allows to specify both, the desired frequency and phase characteristics of filters.
|
|
|
GPU Raytracer for OSG
Kantor, Jiří ; Horváth, Zsolt (referee) ; Starka, Tomáš (advisor)
This work describes creation of a simple raytracer for OpenSceneGraph, which performs its operations on the graphics card. Things, that needed to be done in OpenSceneGraph in order to pass data to the GPU and also several ray-triangle intersection methods, are described in this work.
|
guest ::
