|
|
Creating a Python-based Automated System for Recognizing Emotions from Facial Expressions.
Zima, Samuel ; Malik, Aamir Saeed (referee) ; Hussain, Yasir (advisor)
Táto práca skúma rozpoznávanie výrazu tváre (angl. facial expression recognition - FER) pomocou hlbokého učenia so zameraním na použitie v zariadeniach s obmedzenou pamäťou a výpočtovými zdrojmi. Začína výskumom emócií a výrazov tváre z psychologického, biologického a sociologického hľadiska. Jadro výskumu tvorí návrh a implementácia automatizovaného systému pre FER s použitím súboru dát FER-2013. Tento systém využíva prispôsobenú architektúru SqueezeNet rozšírenú o jednoduchý obchvat, vrstvy náhodného odpadu neurónov a vrstvy dávkovej normalizácie. Tento systém dosahuje na súbore dát FER-2013 presnosť 66,37 %. Pre porovnávaciu analýzu sa tento model porovnal s upravenou architektúrou VGG16, ktorá dosiahla presnosť 65,09 %. Táto práca poskytuje cenné poznatky o vývoji menších, efektívnejších modelov strojového učenia pre FER, ktoré sú použiteľné pre široké spektrum zariadení vrátane nízkovýkonných procesorov a vstavaných zariadení.
|
|
| |
|
|
Design and implementation of a universal data logger
Vokál, Tomáš ; Kovář, Jiří (referee) ; Houška, Pavel (advisor)
Subject of this thesis is to design universal datalogger while using .NET Micro Framewrok. The design was tested on a developing kit EMX produced by GHI Electronics. At the beginning a problematic of .NET Micro Framework is introduced to a reader. After that architecture of project and mechanism of controlling device are described.
|
|
|
Vision Transformery pre vstavané platformy
Nemčeková, Barbora
This work focuses on investigation of Vision Transformers for the task of image classification, their optimization and deployment on selected embedded devices. Until now, convolutional neural networks have been used for image classification on the selected embedded devices, but with the revolution in natural language processing, there has been an interest in investigating transformers for computer vision tasks as well. The work experiments with different kinds of model quantization methods, such as int8 quantization, int16x8 quantization, dynamic quantization, and SmoothQuant method. The results show that not all transformers for computer vision can be quantized with sufficient accuracy, even when using the specialized SmoothQuant method. It also turned out that the quantized transformer model cannot be accelerated on the NPUs of selected devices. From the investigated factors, such as accuracy after model optimization, latency and memory usage on the embedded device, it emerged that for the task of image classification and model deployment on embedded devices, convolutional neural networks still outperform transformer models.
|
|
|
Detection and Classification of Vehicles for Embedded Platforms
Skaloš, Patrik ; Hradiš, Michal (referee) ; Špaňhel, Jakub (advisor)
Táto práca hodnotí kompromisy rýchlosti a presnosti najmodernejších detektorov objektov YOLOv8 pre detekciu vozidiel v snímkoch z monitorovacích kamier na vstatných a nízkovýkonných zariadeniach. Modely YOLOv8 rôznych veľkostí, vrátane jedného s efektívnou sieťou MobileNetV2 na extrakciu príznakov a modelu YOLOv8-femto s menej ako \num{60000} parametrami, boli testované na šiestich zariadeniach, vrátane troch vstavaných platforiem z rodiny NVIDIA Jetson a počítačom Raspberry Pi 4B s nízkou výpočtovou silou. V práci boli zohľadnené rôzne faktory ovplyvňujúce výkonnosť modelov, ako napríklad ich kvantizácia, rozlíšenia vstupu, inferenčné knižnice a veľkosti dávok počas inferencie. Táto štúdia poskytuje užitočné informácie k vývoju a nasadeniu detektorov vozidiel na širokú škálu zariadení, od nízkovýkonných procesorov po špecializované vstavané platformy.
|
|
|
The Progress run-time architecture
Pop, Tomáš
This thesis is a part of a bigger research vision called Progress which aims at providing component based techniques for the development of real-time embedded systems. Progress introduces the concept of a virtual node in order to increase the effectiveness of constructed systems and improve hardware abstraction. The thesis starts research of the runtime structures of the Progress component model. The thesis aims at identifying necessary questions about the runtime internal structure of virtual nodes and about the supporting mechanisms needed to run virtual nodes on destination hardware. A part of this thesis is also a sample implementation of the virtual node runtime environment covering local and Ethernet communication, event driven and timer driven tasks, and multiple computational nodes.
|
|
|
The Progress run-time architecture
Pop, Tomáš
This thesis is a part of a bigger research vision called Progress which aims at providing component based techniques for the development of real-time embedded systems. Progress introduces the concept of a virtual node in order to increase the effectiveness of constructed systems and improve hardware abstraction. The thesis starts research of the runtime structures of the Progress component model. The thesis aims at identifying necessary questions about the runtime internal structure of virtual nodes and about the supporting mechanisms needed to run virtual nodes on destination hardware. A part of this thesis is also a sample implementation of the virtual node runtime environment covering local and Ethernet communication, event driven and timer driven tasks, and multiple computational nodes.
|
|
| |
|
|
Design and implementation of a universal data logger
Vokál, Tomáš ; Kovář, Jiří (referee) ; Houška, Pavel (advisor)
Subject of this thesis is to design universal datalogger while using .NET Micro Framewrok. The design was tested on a developing kit EMX produced by GHI Electronics. At the beginning a problematic of .NET Micro Framework is introduced to a reader. After that architecture of project and mechanism of controlling device are described.
|
guest ::
