|
|
Vyhodnocení směru pohledu osoby
Virgala, Ján ; Odstrčilík, Jan (referee) ; Žalud, Luděk (advisor)
The aim of this bachelor thesis is to understand the problematics of eye tracking. Many current and historical eye tracking methods are described in this work as long as the main principles of their function. This work also informs about an application of the eye tracking systems in many fields of everyday life. This work contains the description of the realized programme and the main principles of its function. An analysis of the practical test performed on the group of volunteers and system-function-analysis are also included.
|
|
| |
|
|
Deep Learning for Object Detection
Pitoňák, Radoslav ; Dobeš, Petr (referee) ; Teuer, Lukáš (advisor)
This thesis analyzes different object detection methods which are based on deep neural networks. In the beginning, the convolutional neural networks are described and commonly used object detection methods are compared. In the following parts, the proposal and implementation of the object detection model trained on the specific dataset are described. In conclusion, the achieved results of this model are discussed and compared with the results of other methods.
|
|
|
System for autonomous racetrack mapping
Soboňa, Tomáš ; Gábrlík, Petr (referee) ; Kopečný, Lukáš (advisor)
The focus of this thesis is to theoretically design, describe, implement and verify thefunctionality of the selected concept for race track mapping. The theoretical part ofthe thesis describes the ORB-SLAM2 algorithm for vehicle localization. It then furtherdescribes the format of the map - occupancy grid and the method of its creation. Suchmap should be in a suitable format for use by other trajectory planning systems. Severalcameras, as well as computer units, are described in this part, and based on parametersand tests, the most suitable ones are selected. The thesis also proposes the architectureof the mapping system, it describes the individual units that make up the system, aswell as what is exchanged between the units, and in what format the system output issent. The individual parts of the system are first tested separately and subsequently thesystem is tested as a whole. Finally, the achieved results are evaluated as well as thepossibilities for further expansion.
|
|
|
Machine Understanding of Maps and Optimal Trajectory Calculation
Kršák, Pavol ; Burian, František (referee) ; Chromý, Adam (advisor)
The teasis deals with the creation of software tools, which use human element to extracts data necessary for navigation in the field of raster maps for orienteering. To achieve this goal, it is important to segment individual sections of the map according to symbols on the map. These symbols are illustrated on the map with one colour or combinations of colours, which create patterns. Because of identification of individual symbols, the work deals with segmentation according to colours and pattern matching. To navigate in the field is important information on height differences on the ground , for this reason the work discusses the position of contour identification and determination of their heights. Combining terrain data with data about height differences will create a 3D color models in XNA. To navigate in such a terrain that is used A* algorithm , in which the test several heuristic model defined distance. The A* algorithm then provides the optimum route according to specified criteria .The resulting application conveys the necessary tools to analyze raster maps orienteering , these tools are the manual and the automatic nature. The work includes detailed operating instructions of developed application.
|
|
|
Automatic Graphics Tool Selection Using Freehand Sketches
Harman, Richard ; Bobák, Petr (referee) ; Čadík, Martin (advisor)
Táto práca je zameraná na klasifikáciu skíc reprezentujúcich telestračné nástroje. Táto klasifikácia bola vytvorená pre telestračnú aplikáciu ako súčasť sekcie na kreslenie voľnou rukou. Po vybratí skíc reprezentujúcich nástroje, zozberaní dát za pomoci aplikácie, ktorú som vytvoril na tento účel, som vytvoril dva datasety. Prvý dataset bol použitý na trénovanie konvolučnej neurónovej siete na klasifikáciu skíc. Druhý dataset bol použitý na trénovanie segmentačnej neurónovej siete pre rozlíšenie línií a elíps v skici. Obe siete boli implementované do aplikácie na ukážku ich funkcionality v reálnom čase a zároveň pre experimentovanie a zisťovanie ich presnosti. Táto aplikácie tiež obsahuje proces dodatočného spracovania, vďaka ktorému vie reprodukovať reprezentácie telestračných nástrojov zo skíc.
|
|
|
Artificial Intelligence for Video Sonification
Dobrocký, Filip ; Burget, Radim (referee) ; Říha, Kamil (advisor)
This thesis deals with the topic of video sonification – the transformation of image into sound. It aims to use state-of-the-art techniques of computer vision based on artificial intelligence to create a system capable of algorithmic sound creation applicable in the art context. The focus is put on the fields of sound art, algorithmic composition and generative music. The thesis includes an implementation of a modular sonification system which utilizes the modern object detector YOLOv7 along with a multiple object tracking algorithm (implemented in the library Norfair), built using the programming language Python. The fundementals of the system lie in systematic assignment of sound objects to objects tracked in the video. The sound creation relies on the SuperCollider platform using the Python API Supriya, incorporating various methods of sound synthesis along with a programmatically created sound database.
|
|
|
Detection and Recognition of Drone Movement in Video
Lapšanský, Simon ; Sakin, Martin (referee) ; Drahanský, Martin (advisor)
S nárastom dostupnosti dronov, narástlo aj riziko ich využívania na nelegálne aktivity. Na základe týchto rizík bola navrhnutá metóda na ich detekciu a následnú klasifikáciu aplikovateľnú v reálnom čase. Navrhovaný prístup využíva metódu odčítania pozadia, slúžiacu na detekciu objektov, zatiaľ čo klasifikácia je dosiahnutá pomocou hlbokého učenia. MOG2 využíva metódu zmiešaného Gaussovho modelu, ktorý slúži na odčítanie pozadia, za účelom detekcie objektov. YOLOv5 model pracujúci s neurónovými sieťami je využitý na následnú klasifikáciu detegovaných objektov. Implementácia vytvára spôsob detekcie a klasifikácie dronov s využitím procesora dosahujúca výsledky postačujúce na aplikovanie detekcie a klasifikácie dronov v reálnom čase. Metóda vyhodnocujúca záznam v rozlíšení 1080p, využívajúca procesor Intel i5-7600K dosahovala v priemere 16 snímiek za sekundu, počas detekcie jedného objektu v snímke.
|
|
|
Displaying Virtual Graphics and Animations for Sports Analysis
Očenáš, Daniel Miloslav ; Zemčík, Pavel (referee) ; Beran, Vítězslav (advisor)
The aim of this bachelor thesis is to create graphic tools used for the analysis of video recordings of sports events, especially football matches. The user can use the tools to highlight selected football players, which are further tracked and highlighted with three-dimensional objects rendered into video scene. Secondly, user can manually create pointed arrows representing player or football movement. Three-dimensional objects are created by perspective distortion corresponding with the scene taken by camera. The theoretical part focuses on the proper rendering of graphical objects into calibrated scene. The practical part focuses on implementation and visualization of mentioned tools by using OpenGL library.
|
|
|
Object Detection in the Laser Scans Using Convolutional Neural Networks
Marko, Peter ; Beran, Vítězslav (referee) ; Veľas, Martin (advisor)
This thesis is aimed at detection of lines of horizontal road markings from a point cloud, which was obtained using mobile laser mapping. The system works interactively in cooperation with user, which marks the beginning of the traffic line. The program gradually detects the remaining parts of the traffic line and creates its vector representation. Initially, a point cloud is projected into a horizontal plane, crating a 2D image that is segmented by a U-Net convolutional neural network. Segmentation marks one traffic line. Segmentation is converted to a polyline, which can be used in a geo-information system. During testing, the U-Net achieved a segmentation accuracy of 98.8\%, a specificity of 99.5\% and a sensitivity of 72.9\%. The estimated polyline reached an average deviation of 1.8cm.
|
guest ::
