|
|
Automated Generation of Realistic Terrain Using Machine Learning Techniques
Střelský, Jakub ; Surynek, Pavel (advisor) ; Holan, Tomáš (referee)
Title: Automated Generation of Realistic Terrain Using Machine Learning Tech- niques Author: Jakub Střelský Department: Department of Theoretical Computer Science and Mathematical Logic Supervisor: RNDr. Pavel Surynek, Ph.D., Department of Theoretical Computer Science and Mathematical Logic Abstract: Artificial terrain is important component of computer games, simulati- ons and films. Manual terrain creation can be arduous process, hence automa- tization of this process would be convenient in many cases. Thanks to current advances in employing artifical neural networks on various generative tasks, the possibility of generating terrain using artificial neural networks should be investi- gated. We will focus on Generative Adversarial Networks as it is one of the most successful content generation method, and we will adjust this method to the task of artificial terrain generation. Resulting model is capable of generating realistic terrain based on raster sketch given by user and allows interactive modelling. Disadvantage of the model is it's requirement of a lot of training data. However, thanks to global elevation datasets providing us with more than enough training data, the model could be useful in certain applications. Keywords: procedural generation, terrain, neural networks, deep learning 1
|
|
|
Creating a database of audio recordings with artificial noise in an anechoic chamber
Hájek, Vojtěch ; Povoda, Lukáš (referee) ; Harár, Pavol (advisor)
This bachelor thesis deals with theory of creating the database of sound records and subsequent creating the database of speech records in the anechoic chamber. Database was created as training dataset for learning process of the artificial neural network, which will be able to separate the speech from background noise. Therefore as the part of the database there are also the recordings of various types of noise that will be used as background noise for the voice recordings. The dataset contains records taken from 18 speakers aged from 16 to 76 years. Half of the speakers were men, half women. Database contains 405 records of speach of average length 46,7 secons and total length 315 minutes. By combining each speech record with each noise record at three levels of signal-to-noise ratio was created 7290 mixed records.
|
|
|
Image annotation using deep learning
Zarapina, Natalya ; Rajnoha, Martin (referee) ; Burget, Radim (advisor)
This semester thesis describes the design and implementation of the client-server program for classification and localization of certain elements which are present in provided images. This program loads a set of images and use deep learning, especially deep convolution neural network perform a classification. First part describes the architecture, basic principles of operations in convolution network and chosen machine learning algorithms for classification. Second part contains a description of created program.
|
|
| |
|
| |
|
|
Deep Learning for Text Classification
Kolařík, Martin ; Harár, Pavol (referee) ; Povoda, Lukáš (advisor)
Thesis focuses on analysis of contemporary machine learning methods used for text classification based on emotion and testing several deep neural nework architectures. Outcome of this thesis is a neural network architecture, which is tuned for using with text data and which had the best result of 79,94 percent. Proposed method is language independent and it doesn’t require as precisely classified training datasets as current methods. Training and testing datasets were consisted of short amateur movie reviews in Czech and in English. Thesis contains also overview of theoretical basics for convolutional neural networks and history of neural networks and language processing Scripts were written in Python, neural networks were simulated using Keras library and Theano framework. We used CUDA for better performance.
|
|
|
Image segmentation using deeplearning methods
Lukačovič, Martin ; Burget, Radim (referee) ; Mašek, Jan (advisor)
This thesis deals with the current methods of semantic segmentation using deep learning. Other approaches of neaural networks in the area of deep learning are also discussed. It contains historical solutions of neural networks, their development, and basic principle. Convolutional neural networks are nowadays the most preferable networks in solving tasks as detection, classification, and image segmentation. The functionality was verified on a freely available environment based on conditional random fields as recurrent neural networks and compered with the deep convolutional neural networks using conditional random fields as postprocess. The latter mentioned method has become the basis for training of new models on two different datasets. There are various enviroments used to implement neural networks using deep learning, which offer diverse perform possibilities. For demonstration purposes a Python application leveraging the BVLC\,/\,Caffe framework was created. The best achieved accuracy of a trained model for clothing segmentation is 50,74\,\% and 68,52\,\% for segmentation of VOC objects. The application aims to allow interaction with image segmentation based on trained models.
|
|
|
Audio noise reduction using deep neural networks
Talár, Ondřej ; Galáž, Zoltán (referee) ; Harár, Pavol (advisor)
The thesis focuses on the use of deep recurrent neural network, architecture Long Short-Term Memory for robust denoising of audio signal. LSTM is currently very attractive due to its characteristics to remember previous weights, or edit them not only according to the used algorithms, but also by examining changes in neighboring cells. The work describes the selection of the initial dataset and used noise along with the creation of optimal test data. For creation of the training network is selected KERAS framework for Python and are explored and discussed possible candidates for viable solutions.
|
|
|
Automatic Machine Learning Methods for Multimedia Data Analysis
Mašek, Jan ; Chromý, Erik (referee) ; Vozňák, Miroslav (referee) ; Burget, Radim (advisor)
The quality and efficient processing of increasing amount of multimedia data is nowadays becoming increasingly needed to obtain some knowledge of this data. The thesis deals with a research, implementation, optimization and the experimental verification of automatic machine learning methods for multimedia data analysis. Created approach achieves higher accuracy in comparison with common methods, when applied on selected examples. Selected results were published in journals with impact factor [1, 2]. For these reasons special parallel computing methods were created in this work. These methods use massively parallel hardware to save electric energy and computing time and for achieving better result while solving problems. Computations which usually take days can be computed in minutes using new optimized methods. The functionality of created methods was verified on selected problems: artery detection from ultrasound images with further classifying of artery disease, the buildings detection from aerial images for obtaining geographical coordinates, the detection of materials contained in meteorite from CT images, the processing of huge databases of structured data, the classification of metallurgical materials with using laser induced breakdown spectroscopy and the automatic classification of emotions from texts.
|
|
|
Deep Neural Networks for Person Identification
Duban, Michal ; Herout, Adam (referee) ; Hradiš, Michal (advisor)
This master's thesis deals with design and implementation of convolutional neural networks used in person re-identification. Implemented convolutional neural networks were tested on two datasets CUHK01 a CUHK03. Results, comparable with state of the art methods were acheved on these datasets. Designed networks were implemented in Caffe framework.
|
guest ::
