prikaz prve stranice dokumenta Monokularna estimacija dubine scene dubokim neuronskim mrežama
No public access
master's thesis
Monokularna estimacija dubine scene dubokim neuronskim mrežama
2021. urn:nbn:hr:168:085516
Gršković, Zvonimir
University of Zagreb
Faculty of Electrical Engineering and Computing

Cite this document

Gršković, Z. (2021). Monokularna estimacija dubine scene dubokim neuronskim mrežama (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:085516

Gršković, Zvonimir. "Monokularna estimacija dubine scene dubokim neuronskim mrežama." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:085516

Gršković, Zvonimir. "Monokularna estimacija dubine scene dubokim neuronskim mrežama." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:085516

Gršković, Z. (2021). 'Monokularna estimacija dubine scene dubokim neuronskim mrežama', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 11 April 2024, https://urn.nsk.hr/urn:nbn:hr:168:085516

Gršković Z. Monokularna estimacija dubine scene dubokim neuronskim mrežama [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2021 [cited 2024 April 11] Available at: https://urn.nsk.hr/urn:nbn:hr:168:085516

Z. Gršković, "Monokularna estimacija dubine scene dubokim neuronskim mrežama", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:168:085516

Please login to the repository to save this object to your list.
Metadata
TitleMonokularna estimacija dubine scene dubokim neuronskim mrežama
Title (english)Monocular scene depth estimation using deep neural networks
AuthorZvonimir Gršković
MentorIvan Marković (mentor)
Committee memberIvan Marković (predsjednik povjerenstva)
Committee memberIvan Petrović (član povjerenstva)
Committee memberBranimir Novoselnik (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2021-02-19, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Abstract
Estimacija dubine važan je izazov u robotici i autonomnoj vožnji. Kada bi se dubina mogla estimirati samo pomoću kamere, mogao bi se izbjeći problem ugradnje i kalibracije dodatnih senzora za tu namjenu. Svake se godine razvijaju nove arhitekture dubokih neuronskih mreža te se upravo pomoću njih može estimirati dubina iz samo jedne slike. Takve se duboke mreže moraju trenirati na skupovima podataka, čije je prikupljanje veliki izazov. U radu su obrađeni problemi treniranja neuronskih mreža te je prikazana njihova točnost estimacije dubine.
Abstract (english)
Depth estimation is an important challenge in robotics and autonomous driving. If depth could be estimated only using a camera, the problem of implementation and calibration of additional sensors could be avoided. Every year new neural network architectures are developed and using those architectures depths could be estimated from one image. Those neural networks are trained on datasets which can be difficult to acquire. This thesis deals with challenges of neural network training and shows accuracy of their depth estimation.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:085516
Study programmeTitle: Electrical Engineering and Information Technology
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-13 12:22:21