Hlavní stránka > Výsledky hledání: keywords:"sports pose"
Hledej:
Tipy pro vyhledávaní :: Rozšířené hledání
Hledej ve sbírkách:
Seřadit podle: Zobrazit výsledky: Výstupní formát:
Národní úložiště šedé literatury Nalezeno 2 záznamů.  Hledání trvalo 0.00 vteřin. 
Mobile User Interface for Comparing Sports Pose Photos
 Hurbánková, Nicol ; Zemčík, Pavel (oponent) ; Herout, Adam (vedoucí práce)
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.
Úplný záznam
Self-Supervised Learning for Recognition of Sports Poses in Image
 Konečný, Daniel ; Beran, Vítězslav (oponent) ; Herout, Adam (vedoucí práce)
The goal of this thesis is to recognize sports poses in image data with a self-supervised learning approach to achieve high classification accuracy even with a low number of annotated samples. Self-supervision is obtained by using images of the same scene from multiple viewpoints at identical and different times. A convolutional neural network trained with triplet loss learns embedding vectors of sports poses and a dense neural network classifies them. The proposed self-supervised model achieves classification accuracy higher by 30-40 % than a supervised model when there are only tens or ones of annotated training samples from each class. The main contributions of this thesis are a set of semi-automatic tools to prepare a dataset for the specific training process, two datasets with sets of labels for classification, and implemented models for specific self-supervised learning. The results show that self-supervised learning is a meaningful approach for solving classification problems with very few labeled samples.
Úplný záznam

Chcete být upozorněni, pokud se objeví nové záznamy odpovídající tomuto dotazu?
Přihlásit se k odběru RSS.
Digitální repozitář :: Hledej :: Nový záznam :: Personalizace :: Nápověda
Powered by Invenio v1.1.2
Spravuje admin.invenio@techlib.cz
Tato stránka je dostupná také v následujících jazycích:
Česky  English