prikaz prve stranice dokumenta Semantička segmentacija slika prometnica generiranih CARLA simulatorom
  Download
PDF 2.54 MB
master's thesis
Semantička segmentacija slika prometnica generiranih CARLA simulatorom
Pula: University of Pula, 2023. urn:nbn:hr:137:228791
Aničić, Simon
University of Pula
Faculty of Informatics in Pula

Institutional repository: Digital Repository Juraj Dobrila University of Pula

Cite this document

Aničić, S. (2023). Semantička segmentacija slika prometnica generiranih CARLA simulatorom (Master's thesis). Pula: University of Pula. Retrieved from https://urn.nsk.hr/urn:nbn:hr:137:228791

Aničić, Simon. "Semantička segmentacija slika prometnica generiranih CARLA simulatorom." Master's thesis, University of Pula, 2023. https://urn.nsk.hr/urn:nbn:hr:137:228791

Aničić, Simon. "Semantička segmentacija slika prometnica generiranih CARLA simulatorom." Master's thesis, University of Pula, 2023. https://urn.nsk.hr/urn:nbn:hr:137:228791

Aničić, S. (2023). 'Semantička segmentacija slika prometnica generiranih CARLA simulatorom', Master's thesis, University of Pula, accessed 29 April 2024, https://urn.nsk.hr/urn:nbn:hr:137:228791

Aničić S. Semantička segmentacija slika prometnica generiranih CARLA simulatorom [Master's thesis]. Pula: University of Pula; 2023 [cited 2024 April 29] Available at: https://urn.nsk.hr/urn:nbn:hr:137:228791

S. Aničić, "Semantička segmentacija slika prometnica generiranih CARLA simulatorom", Master's thesis, University of Pula, Pula, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:137:228791

Metadata
TitleSemantička segmentacija slika prometnica generiranih CARLA simulatorom
Title (english)Semantic segmentation of traffic images generated by CARLA simulator
AuthorSimon Aničić
MentorGoran Oreški (mentor)
Committee memberNikola Tanković (predsjednik povjerenstva)
Committee memberGoran Oreški (član povjerenstva)
Committee memberTihomir Orehovački (član povjerenstva)
GranterUniversity of Pula
(Faculty of Informatics in Pula)
Pula
Defense date and country2023-09-12, Croatia
Scientific / art field,
discipline and subdiscipline		SOCIAL SCIENCES
Information and Communication Sciences
Information Systems and Information Science
Abstract
Rad se bavi proučavanjem, usporedbom i analizom često korištenih funkcija gubitaka i jednostavnih tehnika za augmentaciju podataka koje se mogu primijeniti na rješavanje problema semantičke segmentacije u autonomnoj vožnji. Rad najprije daje strukturirani pregled teorije na kojoj su bazirani znanstveni radovi, a kako bi se detaljnije razumio proces treniranja modela te objasnili korišteni pojmovi i različite tehnike. Drugi dio rada odnosi se na znanstvene radove u kojima se detaljno prikazuje korištenje različitih funkcija gubitaka te tehnika za augmentaciju podataka i njihovih ostvarenih rezultata na različitim modelima. Rezultati ovog rada doprinose svima koji žele naučiti više o spomenutim temama te su u radu dane i preporuke za korištenje funkcija gubitaka i tehnika za augmentaciju podataka u kombinaciji s predtreniranim modelima, ovisno o zadatku koji se nastoji ostvariti njihovim korištenjem.
Abstract (english)
This thesis deals with the study, comparison and analysis of commonly used loss functions and simple data augmentation techniques that can be applied to solving semantic segmentation problems in autonomous driving. The thesis first gives a structured overview of the theory on which the scientific papers are based, in order to understand the model training process in more detail and to explain the concepts and different techniques used. The second part of the thesis refers to scientific papers in which the use of different loss functions and techniques for data augmentation and their achieved results on different pre-trained models are presented in detail. The results of this work contribute to everyone who wants to learn more about the mentioned topics, and recommendations for the use of loss functions and techniques for data augmentation in combination with pre-trained models are given within the thesis, depending on the task that is sought to be achieved by using them.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:137:228791
Study programmeTitle: Master in Informatics; specializations in: Master in Informatics, Master in Informatics, specialisation: teaching, Master in Informatics, Master in Informatics, specialisation: teaching
Course: Master in Informatics
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra informatike (magistar/magistra informatike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2023-09-13 15:13:18