prikaz prve stranice dokumenta Učenje 3D reprezentacije objekata u visokoj rezoluciji
No public access
undergraduate thesis
Učenje 3D reprezentacije objekata u visokoj rezoluciji
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. urn:nbn:hr:168:634765
Kušina, Luka
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Kušina, L. (2019). Učenje 3D reprezentacije objekata u visokoj rezoluciji (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:634765

Kušina, Luka. "Učenje 3D reprezentacije objekata u visokoj rezoluciji." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. https://urn.nsk.hr/urn:nbn:hr:168:634765

Kušina, Luka. "Učenje 3D reprezentacije objekata u visokoj rezoluciji." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. https://urn.nsk.hr/urn:nbn:hr:168:634765

Kušina, L. (2019). 'Učenje 3D reprezentacije objekata u visokoj rezoluciji', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 26 November 2024, https://urn.nsk.hr/urn:nbn:hr:168:634765

Kušina L. Učenje 3D reprezentacije objekata u visokoj rezoluciji [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2019 [cited 2024 November 26] Available at: https://urn.nsk.hr/urn:nbn:hr:168:634765

L. Kušina, "Učenje 3D reprezentacije objekata u visokoj rezoluciji", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2019. Available at: https://urn.nsk.hr/urn:nbn:hr:168:634765

Metadata
TitleUčenje 3D reprezentacije objekata u visokoj rezoluciji
Title (english)Learning 3D Representations at High Resolution
AuthorLuka Kušina
MentorTomislav Pribanić (mentor)
Committee memberTomislav Pribanić (predsjednik povjerenstva)
Committee memberKristijan Bartol (član povjerenstva)
Committee memberDavid Bojanić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2019-09-02, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U sklopu završnog rada predstavili smo dvije podatkovne strukture za pohranjivanje 3D objekata – Voksel i Octree te predstavljene su prednosti i mane korištenje obje podatkovne strukture. Nakon kratkog predstavljanja softverske podrške i skupa podataka nad kojiim se vrši treniranje mreže, opisan je puni proces treniranja konvolucijske neuronske mreže za klasifikaciju 3D objekata sa vokselima kao ulaznim podatcima. Potom su rezultati testiranja i validacije dobivene mreže uspoređeni sa rezultatima OctNet neuronske mreže te je donesen zaključak o učinkovitosti navedenih podatkovnih struktura.
Abstract (english)
We present two different data structure representationes of 3D objects – Voxels and Octrees, and detail the pros and cons of both representations. After a short introduction of the software used in the scope of the project and the dataset used as the training and test dataset in our deep learning neural network, we present the full process of training a convolutional neural network for 3D classification with the voxel representations of the original dataset. Afterwards, the training and validation results are taken into consideration and a conclusion on the effectiveness of both 3D representations is drawn.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:634765
Study programmeTitle: Computing
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2020-01-07 19:35:13