20.500.12556/DKUM-88841 Določanje uspešnosti počepa s pomočjo strojnega vida Determining squat depth with the help of machine vision V tem diplomskem delu bomo predstavili področje strojnega učenja, bolj specifično področje globokega učenja. V teoretičnem delu bomo prikazali, kako se je strojno učenje do sedaj že uporabljalo v športu, kako strojno in globoko učenje delujeta ter kako poteka proces učenja konvolucijskih nevronskih mrež. V praktičnem delu bomo ustvarili svojo učno množico in nato z algoritmom, ki je zasnovan na konvolucijskih nevronskih mrežah, ustvarili model, ki je zmožen določati uspešnost počepa po pravilih zveze za Powerlifting. In this diploma thesis, we will delve into the field of machine learning, specifically focusing on the area of deep learning. In the theoretical part, we will examine how machine learning has been until now used in sports and as well as how machine learning, deep learning, and the process of training convolutional neural networks work. In the practical part, we will create our own training dataset and use an algorithm based on convolutional neural networks to create a model capable of assessing the performance of a squat according to the rules of the International Powerlifting Federation. strojni vid globoko učenje konvolucijske nevronske mreže powerlifting machine learning deep learning convolutional neural networks powerlifting true true false [N. Graj] Slovenski jezik Angleški jezik Diplomsko delo/naloga 2024-05-29 17:43:20 2024-09-19 15:34:08 2024-12-23 09:59:43 0000-00-00 00:00:00 2024 0 Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.8796.88(043.2) 220206083 VS_Graj_Nejc_2024.pdf VS_Graj_Nejc_2024.pdf 1 0F41AF0095760FE4113B85DCE0672521 8407b3597f4fd59a7549537d4fc9c7c6ce279b2cd0a5bd1bfa6b6455af2a99db 1aa5d3ad-3979-11ef-ad11-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=181078 Fakulteta za elektrotehniko, računalništvo in informatiko 0 0 0