Sustav za procjenu poze sportaša temeljen na dubokom učenju

master's thesis

• Title (english): Athletes Pose Estimation Based on Deep Learning
• Author: Pino Pavlić
• Mentor: Vladimir Čeperić (mentor)
• Committee member: Vladimir Čeperić (predsjednik povjerenstva)
• Committee member: Zoran Kalafatić (član povjerenstva)
• Committee member: Tomislav Marković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-17, Croatia

Abstract

Ovaj rad bavi se detaljnom analizom ispravnosti pokreta šuta slobodnog bacanja u košarci. Objašnjava osnove modela za procjenu položaja tijela te proces dobivanja podataka o koordinatama ključnih točaka tijela iz videozapisa. Opisuje proces prilagođavanja videozapisa na isti broj okvira te detaljnu analizu ključnih dijelova tijela, pri čemu identificira najbitnije značajke koje utječu na kvalitetu šuta. Opisuje model za klasifikaciju ispravnosti pokreta slobodnog bacanja koristeći optimizirane hiperparametre i odabrane značajke. Rad pokazuje kako kombinacija temeljite analize podataka i naprednih metoda strojnog učenja može postići visoku točnost u klasifikaciji složenih pokreta u sportu.

Abstract (english)

This paper deals with a detailed analysis of the correctness of the free throw shot movement in basketball. It explains the basics of the pose estimation model and the process of obtaining data about the coordinates of body key points from a video. It describes the process of adjusting the video to the same number of frames and a detailed analysis of key body parts, identifying the most important features that affect the quality of the shot. Describes a model for classifying free throw motion accuracy using optimized hyperparameters and selected features. The paper shows how a combination of thorough data analysis and advanced machine learning methods can achieve high accuracy in the classification of complex movements in sports.

• Keywords: procjena poze; strojno učenje; računalni vid; sport
• Keywords (english): pose estimation; machine learning; computer vision; sport
• Language: croatian
• URN:NBN: urn:nbn:hr:168:003219
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-10-12 21:48:14