Procjena poze sportaša temeljeno na dubokom učenju

master's thesis

• Title (english): Athletes Pose Estimation Based on Deep Learning
• Author: Anel Hadžimuratagić
• Mentor: Vladimir Čeperić (mentor)
• Committee member: Vladimir Čeperić (predsjednik povjerenstva)
• Committee member: Siniša Šegvić (član povjerenstva)
• Committee member: Mirko Poljak (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Procjena poze je jedan od važnijih problema u području računalnog vida. U ovom
radu upoznali smo se s najvažnijim primjenama procjene poze u sportu. Opisali smo
što je duboko učenje te smo se upoznali s njegovim ključnim dijelovima. Prikazali smo
glavne dijelove implementacije modela te predstavili dosadašnja istraživanja. Predstavili
smo dva pristupa: odozgo prema dolje i odozdo prema gore koji smo koristili za
implementaciju ovog problema. Detaljno smo opisali skup podataka Leeds Sport Pose
te prikazali i opisali dobivene rezultate na tom skupu. Na kraju predložili moguća
poboljšanja i nadogradnju modela.

Abstract (english)

Pose estimation is one of the biggest problems in computer vision area. In this
paper you were introduced with the most important usage of pose estimation in sports.
Deep learning was described along side its key features. Main parts of the model
implementation were presented, as well as the previous researches. Two approaches
were shown: top-up approach and bottom-up approach which we used for the implementation.
In addition, we have explained in detail the Leeds Sport Pose dataset and
presented the results of the work. In the end, we suggested possible improvements and
upgrades of our problem.

• Keywords: duboko učenje; procjena poze; Leeds Sport Pose; sport; ključne točke; konvolucijske mreže; binarna klasifikacija
• Keywords (english): deep learning; pose estimation; Leeds Sport Pose; sport; keypoints; convolutional network; binary classification
• Language: croatian
• URN:NBN: urn:nbn:hr:168:091818
• 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: Closed access
• Created on: 2023-01-13 12:23:49