prikaz prve stranice dokumenta Određivanje dubine scene temeljeno na dubokom učenju
No public access
master's thesis
Određivanje dubine scene temeljeno na dubokom učenju
2019. urn:nbn:hr:168:360029
Vernier, Marin
University of Zagreb
Faculty of Electrical Engineering and Computing

Cite this document

Vernier, M. (2019). Određivanje dubine scene temeljeno na dubokom učenju (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:360029

Vernier, Marin. "Određivanje dubine scene temeljeno na dubokom učenju." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. https://urn.nsk.hr/urn:nbn:hr:168:360029

Vernier, Marin. "Određivanje dubine scene temeljeno na dubokom učenju." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2019. https://urn.nsk.hr/urn:nbn:hr:168:360029

Vernier, M. (2019). 'Određivanje dubine scene temeljeno na dubokom učenju', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 24 February 2025, https://urn.nsk.hr/urn:nbn:hr:168:360029

Vernier M. Određivanje dubine scene temeljeno na dubokom učenju [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2019 [cited 2025 February 24] Available at: https://urn.nsk.hr/urn:nbn:hr:168:360029

M. Vernier, "Određivanje dubine scene temeljeno na dubokom učenju", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2019. Available at: https://urn.nsk.hr/urn:nbn:hr:168:360029

Please login to the repository to save this object to your list.
Metadata
TitleOdređivanje dubine scene temeljeno na dubokom učenju
Title (english)Estimation of Scene Depth Based on Deep Learning
AuthorMarin Vernier
MentorTomislav Hrkać (mentor)
Committee memberTomislav Hrkać (predsjednik povjerenstva)
Committee memberSiniša Šegvić (član povjerenstva)
Committee memberZoran Kalafatić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2019-07-04, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Tehnike procjene dubine scene nalaze široku primjenu u raznim područjima i zahtjeva se da budu pouzdane i robusne. U posljednje vrijeme klasične tehnike procjene dubine scene zamijenjene su tehnikama koje se baziraju na dubokom učenju. U ovom radu dan je pregled područja dubokog učenja, stereo računalnog vida te primjene dubokog učenja u procjeni dubine scene. Naučeno je nekoliko modela, primjenom dvije arhitekture mreža, StereoConvNet i DispNet, nad podatkovnim skupovima Foucard i SceneFlow. Nad dobivenim modelima provedena su mjerenja te je dana usporedba performansi.
Abstract (english)
Depth estimation techniques have a broad range of applications among various domains. It is required that these techniques are reliable and robust. Lately, classic depth estimation techniques have been replaced by techniques based on deep learning. This thesis contains an overview of deep learning, stereo computer vision, and the application of deep learning concepts to the depth estimation problem. Several models have been learned, using StereoConvNet and DispNet architectures and Foucard and SceneFlow datasets. An analysis has been done among gathered models, and model performance comparison has been given.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:360029
Study programmeTitle: Computing
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 conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2020-01-07 22:49:04