| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

Show document Help

Title:Rekonstrukcija, upodabljanje in analiza scene z metodo Gaussovega packanja : magistrsko delo
Authors:ID Žagar, Dominik (Author)
ID Kohek, Štefan (Mentor) More about this mentor... New window
ID Strnad, Damjan (Comentor)
Files:.pdf MAG_Zagar_Dominik_2026.pdf (23,49 MB)
MD5: 408D3D2A06E475E71FD42A4F60EB81C5
 
Language:Slovenian
Work type:Master's thesis/paper
Typology:2.09 - Master's Thesis
Organization:FERI - Faculty of Electrical Engineering and Computer Science
Abstract:V tem magistrskem delu smo obravnavali metodo Gaussovega packanja za 3D rekonstrukcijo in analizo scen iz 2D slik. Najprej smo pregledali teoretične osnove sevalnih polj in metode, kot je NeRF, ter primerjali njihove prednosti in slabosti. Metodo Gaussovega packanja smo podrobno opisali kot proces, ki namesto nevronskih mrež uporablja oblake točk za hitro in učinkovito rekonstrukcijo scen. V praktičnem delu magistrskega dela smo razvili aplikacijo, ki poenostavi proces pridobitve vizualizacije oblaka pack iz zajetih slik. Aplikacija vključuje tudi analizo kakovosti rekonstrukcije in interaktivno pregledovanje scen. Rezultati magistrskega dela kažejo, da Gaussovo packanje omogoča hitro in kakovostno rekonstrukcijo, primerno za različna področja, kot sta gradbeništvo in kinematografija. Glavni prednosti metode sta preprostost uporabe in realnočasovna vizualizacija, omejitve pa vključujejo odvisnost od začetne kakovosti rekonstrukcije. Zaključujemo, da je Gaussovo packanje perspektivna metoda za praktične 3D rekonstrukcije z nadaljnjim razvojem predvsem na področju izboljšanja začetne rekonstrukcije.
Keywords:računalniška grafika, sevalna polja, 3D rekonstrukcija, oblaki točk, vizualizacija
Place of publishing:Maribor
Place of performance:Maribor
Publisher:[D. Žagar]
Year of publishing:2025
Number of pages:1 spletni vir (1 datoteka PDF (XI, 53 str.))
PID:20.500.12556/DKUM-97264 New window
UDC:004.925(043.2)
COBISS.SI-ID:275071235 New window
Publication date in DKUM:24.03.2026
Views:178
Downloads:42
Metadata:XML DC-XML DC-RDF
Categories:KTFMB - FERI
:
Copy citation
  
Average score:(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share



Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:CC BY-NC 4.0, Creative Commons Attribution-NonCommercial 4.0 International
Link:http://creativecommons.org/licenses/by-nc/4.0/
Description:A creative commons license that bans commercial use, but the users don’t have to license their derivative works on the same terms.
Licensing start date:25.02.2026

Secondary language

Language:English
Title:Scene reconstruction, visualization and analysis with Gaussian splatting method
Abstract:In this master’s thesis, we addressed the Gaussian splatting method for 3D reconstruction and scene analysis from 2D images. First, we reviewed the theoretical foundations of radiance fields and methods such as NeRF, and compared their advantages and disadvantages. We described Gaussian splatting in detail as a process that, instead of neural networks, uses point clouds for fast and efficient scene reconstruction. In the practical part of the master’s thesis, we developed an application that simplifies the process of obtaining a visualisation of the splat cloud from captured images. The application also includes reconstruction quality analysis and interactive scene inspection. The results of the master’s thesis show that Gaussian splatting enables fast and high-quality reconstruction, suitable for various fields such as construction and cinematography. The main advantages of the method are ease of use and real-time visualisation, while the limitations include dependence on the initial reconstruction quality. We conclude that Gaussian splatting is a promising method for practical 3D reconstructions, with further development focused primarily on improving the initial reconstruction.
Keywords:computer graphics, radiance fields, 3D reconstruction, point clouds, visualization


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica