|
|
Application of Neural Accelerators on Rapsberry PI
Barna, Kristian ; Sekanina, Lukáš (referee) ; Vašíček, Zdeněk (advisor)
The presented bachelor thesis deals with the statistical evaluation of performance for hardward accelerator of deep neural networks. Describes convolutional neural networks along with mathematical calculations. Explains their acceleration and conversion to a format suitable for the Intel Movidius NCS accelerator. 8 hardware platforms and 22 neural network difficulties were compared experimentally. Up to 105-fold improvement was demonstrated in isolated inference of the MobileNetV2 network for the Raspber Pi platform using an accelerator. Performance between the tested platforms was also evaluated from an energy point of view. The application of facial identity demonstrated the conditions during real use. Possible limits of CNN acceleration on power-limited devices (Raspberry Pi) have been uncovered, especially due to improper selection of input image resolution. All measurements were evaluated by statistical procedures.
|
|
|
Deep Learning for Medical Image Analysis
Szöllösi, Albert ; Kodym, Oldřich (referee) ; Španěl, Michal (advisor)
This thesis offers possible solution to automatic 3D dental scan landmark localization. These scans are used in dental crown design and digital orthodontics to make the design process easier using specialized software. Before that, though, the scan has to be annotated for the software to know the positions of the teeth. The annotation process is done manually, which guarantees precision, but takes a lot of time. The result of this work could make said process much simpler by applying deep learning. Landmark localization was implemented using a convolutional neural network.
|
|
|
Recurrent Neural Network for Text Classification
Myška, Vojtěch ; Kolařík, Martin (referee) ; Povoda, Lukáš (advisor)
Thesis deals with the proposal of the neural networks for classification of positive and negative texts. Development took place in the Python programming language. Design of deep neural network models was performed using the Keras high-level API and the TensorFlow numerical computation library. The computations were performed using GPU with support of the CUDA architecture. The final outcome of the thesis is linguistically independent neural network model for classifying texts at character level reaching up to 93,64% accuracy. Training and testing data were provided by multilingual and Yelp databases. The simulations were performed on 1200000 English, 12000 Czech, German and Spanish texts.
|
|
|
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.
|
|
|
Depth-Based Determination of a 3D Hand Position
Ondris, Ladislav ; Tinka, Jan (referee) ; Drahanský, Martin (advisor)
Cílem této práce je určení kostry ruky z hloubkového obrazu a jeho následné využití k rozpoznání statického gesta. Na vstupu je hloubkový obrázek, ve kterém je nejprve detekována ruka pomocí neuronové sítě Tiny YOLOv3. Následně je obrázek zbaven pozadí a z takto předzpracovaného obrázku je určena kostra ruky v podobě 21 klíčových bodů neuronovou sítí JGR-P2O. K rozpoznání gesta z klíčových bodů ruky byla navržena technika, která porovná kostru na vstupu s uživatelem definovanými gesty. Funkcionalita systému byla otestována na vytvořeném datasetu s více než čtyřmi tisíci obrázky.
|
|
|
Automatic Humor Evaluation
Katrňák, Josef ; Ondřej, Karel (referee) ; Dočekal, Martin (advisor)
The aim of this thesis is to create a system for automatic humor evaluation. The system allow to predict humor and category for english input. The main essence is to create a classifier and train the model with the created datasets to get the best possible results. The classifier architecture is based on neural networks. The system also includes a web user interface for communication with the user. The result is a web application linked to a classifier that allows user input to be evaluated and user feedback to be provided.
|
|
|
Deep Learning Algorithms on Embedded Devices
Hadzima, Jaroslav ; Boštík, Ondřej (referee) ; Horák, Karel (advisor)
Táto práca popisuje v súčastnosti široko používané architektúry a modely pre Hlboké Učenie, riešiace úlohu detekcie a klasifikácie objektov vo videu. Dôraz tu bude kladený na ich použiteľnosť na vstavaných zariadeniach. Postupne preberieme kroky a odvôvodňovanie pri výbere najlepšieho vstavaného systému pre našu aplikáciu. Ukážková aplikáci pozostáva hlavne z detekcie vozidiel a detekcie voľných parkovacích miest s využitím algoritmov Hlbokého Učenia. Táto aplikácia umožňuje monitorovať počet vozidiel, nachádzajúcich sa na parkovisku a zároveň rozhodnúť, či sa nachádzajú na prakovacom mieste alebo nie. Následne tu budú prebrané kroky nutné ku konfigurácii zariadenia s dôrazom na optimalizáciu hardvéru pre dosiahnutie čo najväčšej rýchlosti. V ďaľšej časti bude poskytnuté porovnanie vybraných modelov, ktoré budú porovnávané hlavne v kategóriách ako rýchlosť alebo F1 skóre. Najlepší kandidát bude použitý na riešenie našej aplikácie a následné testovanie jej vlastností s názvom Inteligentné parkovisko.
|
|
|
Convolutional neural networks for identification of axial 2D slices in CT data
Vavřinová, Pavlína ; Harabiš, Vratislav (referee) ; Jakubíček, Roman (advisor)
This thesis deals with the classification of axial 2D slices in CT patient’s data into six categories. The sphere of convolutional neural networks was used for this purpose. For a better understanding of this issue, the basics of neural networks and then the principles of deep learning including convolutional neural networks are explained at first. The AlexNet network was specifically selected for the intention of this identification, and it was tested on the created data set after being adaptated. The overall classification success rate was 86% ,after the final adjustments, a slight improvement was achieved and the identification success rate was 87%.
|
|
|
Deep Learning for Image Recognition
Munzar, Milan ; Kolář, Martin (referee) ; Hradiš, Michal (advisor)
Neural networks are one of the state-of-the-art models for machine learning today. One may found them in autonomous robot systems, object and speech recognition, prediction and many others AI tasks. The thesis describes this model and its extension which is used in an object recognition. Then explains an application of a convolutional neural networks(CNNs) in an image recognition on Caltech101 and Cifar10 datasets. Using this exemplar application, the thesis discusses and measures efficiency of techniques used in CNNs. Results show that the convolutional networks without advanced extensions are able to reach a 80\% recognition accuracy on Cifar-10 and a 37\% accuracy on Caltech101.
|
|
|
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 ::
