Metode za procjenu ljudske poze

master's thesis

• Title (english): Methods for Human Pose Estimation
• Author: Danijel Popović
• Mentor: Tomislav Pribanić (mentor)
• Committee member: Tomislav Pribanić (predsjednik povjerenstva)
• Committee member: Marko Subašić (član povjerenstva)
• Committee member: Darko Vasić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-02-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Svakodnevni susret sa sustavima koji koriste metode računalnog vida je neizbježan. Zahvaljujući dubokom učenju problemi računalnog vida ponovno postaju jako popularni. Među jednim od najzanimljivijih jest procjena ljudske poze u 2D/3D. U ovom radu istražene su te potom opisane metode standardnog (klasičnog) pristupa te pristupa koji koriste duboko učenje. Unutar te podjele naglašene su razlike njihovih pristupa i detaljno opisani postupci njihovih najboljih metoda. Implementirane su dvije... More metode za procjenu poze te je napravljena evaluacije provedenih eksperimenata i službenih rješenja. Less

Abstract (english)

Everyday encounter with systems that use computer vision methods is inevitable. Thanks to deep learning, computer vision problems are becoming very popular again. Among one of the most interesting is the estimation of human pose in 2D/3D. In this paper, the methods of the standard (classical) approach and approaches that use deep learning were investigated and then described. Within that division, the differences in their approaches are emphasized and the procedures of their best methods... More are described in detail. Two methods for pose estimation were implemented and an evaluation of the conducted experiments and official solutions was made. Less

• Keywords: procjena ljudske poze u 2D; procjena ljudske poze u 3D
• Keywords (english): human pose estimation in 2D; human pose estimation in 3D
• Language: croatian
• URN:NBN: urn:nbn:hr:168:602105
• 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: 2023-07-13 10:29:15