prikaz prve stranice dokumenta Automatizirano generiranje 3D modela ljudi
  Download
PDF 2.59 MB
master's thesis
Automatizirano generiranje 3D modela ljudi
2023. urn:nbn:hr:200:153635
Orić, Mihaela
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Software Engineering
Chair of Visual Computing

Cite this document

Orić, M. (2023). Automatizirano generiranje 3D modela ljudi (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:153635

Orić, Mihaela. "Automatizirano generiranje 3D modela ljudi." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:153635

Orić, Mihaela. "Automatizirano generiranje 3D modela ljudi." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:153635

Orić, M. (2023). 'Automatizirano generiranje 3D modela ljudi', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 12 May 2024, https://urn.nsk.hr/urn:nbn:hr:200:153635

Orić M. Automatizirano generiranje 3D modela ljudi [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2023 [cited 2024 May 12] Available at: https://urn.nsk.hr/urn:nbn:hr:200:153635

M. Orić, "Automatizirano generiranje 3D modela ljudi", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:200:153635

Please login to the repository to save this object to your list.
Metadata
TitleAutomatizirano generiranje 3D modela ljudi
Title (english)Automated generation of 3D models of humans
AuthorMihaela Orić
MentorČaslav Livada (mentor)
Committee memberKrešimir Nenadić (predsjednik povjerenstva)
Committee memberČaslav Livada (član povjerenstva)
Committee memberRobert Šojo (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Software Engineering)
(Chair of Visual Computing)
Osijek
Defense date and country2023-09-12, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Data Processing
Abstract
Ovaj rad opisao je primjenu sintetičkih podataka za strojno učenje te predstavio nekoliko postojećih sustava za izradu sintetičkih podataka ljudi. Zbog nedostataka tih sustava, npr. nedostupnost korištenja sustava za slobodnu uporabu, predstavljen je novi sustav za kreiranje modela ljudi. Sustav kreira 3D modele ljudi u alatu za 3D modeliranje, Blender, korištenjem javno dostupnih proširenja. Svaki model čovjeka detaljno se uređuje kako bi bio što realističniji. Poseban naglasak stavljen je na uređivanje teksture ljudske kože. Poboljšanje izgleda izrađenih ljudi moguće je napraviti proširenjem baze modela odjeće i kose koji se dodaju na modele. Modificiraju se poza čovjeka i scena u kojoj se nalazi svaki čovjek, te se kao izlaz dobivaju slike scene s određenim anotacijama. Cijeli postupak automatiziran je korištenjem Pythona. Pokretanje i odabir opcija generiranja podataka vrši se putem grafičkog sučelja.
Abstract (english)
This thesis describes the use of synthetic data in machine learning and introduces a few existing solutions for synthetic human data generation. The downfalls of those systems, such as the fact that those systems can not be freely used publicly, gave motivation for introducing a pipeline for generation of human models. Proposed pipeline creates 3D models of people in a 3D modelling tool Blender by using publicly available addons. Each model is edited in detail to achieve a realistc look. Focus is given on texturing the human skin. The look of models could be improved by extending the base of clothes and hair models. The pose of models and the scene they are in is modified and an anotated render is given as an output. The whole pipeline is automated using Python. Generation start and parameters selection are done in a graphical interface.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:153635
Study programmeTitle: Graduate University Study Programme in Computer Engineering
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2023-10-09 08:45:21