|
|
Unsupervised Anomaly Detection in Image
Salvet, Lukáš ; Herout, Adam (referee) ; Juránek, Roman (advisor)
This thesis deals with anomaly detection on industrial products. The main requirement was that the method required as little data with anomalies as possible at the time of construction and that it was easily applicable to different types of products. Neural network that is indirectly taught to find differences between two pictures is designed and described in this thesis. The anomaly detection itself should take place based on the representation of input data in latent space or in combination with a reconstruction loss. Four different method modifications have been designed and tested. The testing was mainly carried out on the MVTec AD dataset, which contains industrial products. Unfortunately the assumption that if the network is taught to look for differences the latent space will be interpreted better was not confirmed. Therefore the method was evaluated in a reconstructive error mode in~which it achieves comparable results with other methods. The result is insufficient for use in practice.
|
|
|
GAN Generated Data for CNN Age Estimation
Venkrbec, Tomáš ; Herout, Adam (referee) ; Hradiš, Michal (advisor)
The goal of this thesis is to implement one of the state-of-the-art methods of generative adversarial networks and to propose its extension to conditional generation. This has been used to generate photorealistic images of human faces with specified characteristics such as age and gender. For this purpose, a highly diverse dataset of over 230,000 samples was created by merging and cleaning existing annotated face datasets. All ages, genders and different ethnic groups are well represented in it. StyleGAN2 generator trained on this dataset achieved a FID of 7.14. The synthetic data ratio was then experimented with during age classifier training. For the test subset of the dataset, the addition of synthetic data achieved a reduction in the mean absolute error from 3.499 years to 3.294 years. For the independent test dataset, a reduction in mean error from 4.012 years to 3.875 years was achieved.
|
|
|
Web App for Q-Sorting
Janů, Michal ; Bartl, Vojtěch (referee) ; Herout, Adam (advisor)
Hlavním cílem této práce je vytvořit a iterativně otestovat uživatelské rozhraní (UI) s uživatelskou zkušeností (UX) pro aplikaci podporující Q-řazení, se zaměřením na třídění karet. Aplikace umožňuje řazení (přesun a výměnu) karet obsahující tvrzení, která jsou řazena na základě určité otázky. Dále jsou podporovány další funkce a elementy, které usnadňují třídění karet a celkovou práci s aplikací. Aplikace například umožňuje resetovat celou tabulku s kartami, automaticky ukládá už částečně seřazenou tabulku a poskytuje "minimap" element s náhledem nad celou tabulkou.
|
|
|
Tracking of Moving Objects in Video
Kroupa, Dominik ; Špaňhel, Jakub (referee) ; Herout, Adam (advisor)
This thesis deals with the issue of multi-target multi-camera tracking, which consists of detection, tracking and re-identification of objects. The framework designed to deal with this problem uses state-of-the art methods followed by post-processing of data retrieved by these methods to improve overall accuracy and object re-identification across cameras. The data obtained by the multi-target multi-camera tracking can be further used in tasks such as crowd behaviour analysis in case of pedestrian tracking, or traffic management in case of vehicle tracking. As part of this solution, this work took part in the new challenge of pedestrian tracking, issued by AI City Challenge for year 2023, with a score of 0.2533 IDF1, resulting in 21st place.
|
|
|
Mobile User Interface for Comparing Sports Pose Photos
Hurbánková, Nicol ; Zemčík, Pavel (referee) ; Herout, Adam (advisor)
The aim of this bachelor thesis is to find the most suitable way of portraying photos for a mobile user interface, which will make it possible to compare images of sports positions, in order to observe the results, whether it is progress or vice versa. The task is to design, later prototype or implement and finally test different photo displays. For the purpose of testing, datasets of photographs of different sports positions but also other phenomena where changes can be observed were collected. With the created display methods, user testing was performed.
|
|
|
Innovative Web App: Smart Queue
Zhukov, Nikita ; Zemčík, Pavel (referee) ; Herout, Adam (advisor)
Cílem této bakalářské práce je vytvořit moderní webovou aplikaci, která umožňuje inteligentní správu front. Výrazně sníží obavy lidí z čekání ve fyzických frontách. Nápad na tuto práci vznikl během pandemie v roce 2020, což vedlo k průzkumu již existujících řešení pro fronty. Aplikace prošla dvěma verzemi implementace a technologickými zásobníky. Finální verze využívá zásobník T3, kde Next JS je přední frontendový framework poháněný Typescriptem a Tailwind CSS. Navíc využívá většinu nejnovějších funkcí, které Next JS poskytuje. Pro nasazení části Frontend je použit Vercel, zatímco Backend je hostován v Google Cloud pomocí Firebase. Při vývoji byla použita celá řada nástrojů Firebase, například Firestore Database, Realtime Database, Authentication a Cloud Functions. Tento článek představuje již existující nástroje pro správu front, moderní přístupy ve vývoji a návrh a implementaci aplikace SmartQueue s nápady na budoucí údržbu.
|
|
|
Web Application for Organizing Sports Tournaments
Tuka, Samuel ; Tesařová, Alena (referee) ; Herout, Adam (advisor)
The main goal of this bachelor’s thesis was to create a web application for organizing sports tournaments. The application allows creating tournaments, managing tournaments, adding players, generating matches, recording results, and sending automatic notifications. The application consists of a server-side implemented using Node.js, Express.js framework and a client-side implemented in React. The data that the application uses are stored in a MySQL database. Match generation is based on a similar system to the Swiss system and the ELO values that express the performance of the players. The result of the work is a functional web application that is responsive and can be used on mobile devices. The contribution of the work is to facilitate the organization of a sports tournament called the University Badminton League. However, the application includes multiple sports and can be beneficial for other organizations that decide to use it. The testing of the application was conducted directly during the organization of the University Badminton League.
|
|
|
Application for Support of Agility
Marková, Natália ; Tesařová, Alena (referee) ; Herout, Adam (advisor)
The goal of this bachelor thesis is to design, develop, and test a smartphone app that allows dog owners and trainers to track and analyze agility training for their dogs. The app uses the Firebase database to store its data and is developed using the cross-platform mobile app development framework Flutter. The dog agility training has been thoroughly analyzed for the development of the app. The design of the app is concentrated on providing a user-friendly interface for documenting training sessions, monitoring progress, and offering statistic evaluation. The evaluation’s results will demonstrate how popular the software is by dog trainers and owners as well as how beneĄcial it is in assisting them in enhancing their agility training methods. Also, the user-friendliness and level of user satisfaction of the app will be assessed. Ultimately, the purpose of this thesis is to demonstrate how mobile technology may improve dog agility training and how the developed app offers dog trainers and owners a useful tool for tracking and enhancing their dogs’ agility performance.
|
|
|
Adding Flight Trajectory of VHF Communication with Transcription
Rajko, Adam ; Herout, Adam (referee) ; Szőke, Igor (advisor)
The aim of this project is to create an application for adding speech communication between the pilot and ATC (Air Traffic Control) to the flight trajectory of the aircraft. The inspiration for this work is the ATCO2 project. Based on the studied mapping frameworks and services providing aircraft position information, an application design has been performed. Based on the design and used technologies, the application was successfully implemented. The resulting functionality of the application was tested on users using a questionnaire.
|
|
|
Experiments with Estimation of Human Pose in Image and Video
Horejš, Michal ; Kocur, Viktor (referee) ; Herout, Adam (advisor)
The detection of phenomena in the image has a wide application in many fields and it is therefore important to constantly develop and improve the detection. This work specifically deals with the problem of detecting sports positions in images and videos. The goal was to experiment with tools for human pose recognition and general phenomenon recognition. During the experiments, three new data sets were created, in which key points of the human body are used. The datasets were then trained on several modeled architectures of convolutional neural networks. The results of the experiments show that the appropriate use of key points can help with the detection of sports positions.
|
guest ::
