Institutional Repository of Vilnius University
Fotografijos informacinės sistemos kūrimas /
Abstract
dc:descriptionThe main goal of the thesis is to create an image sharing site for photographers, with a unique set of functionalities. The thesis tasks were: to compare different artificial intelligence models, to define the unique features to be developed based on similar system analysis, to analyse the advantages and disadvantages of different system architectures, to define the functional and non-functional system requirements, to design the system architecture and database diagrams, to develop the system, test it thoroughly and provide recommendations for improvement. After analysing „Instagram“, „Flickr“ and „500px“ photo sharing systems two unique functionalities were proposed: automatic photo tagging using an accurate artificial intelligence model, and a hierarchy-based dynamic photo search. A comparison of four different models led to the conclusion that „EfficientNet“ architecture is best at classifying photos accurately. The system was developed, performance testing was carried out to identify and correct existing problems, improving the system‘s performance during heavy load, and recommendations were made for further system development.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Česnavičius, Simas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD210646843&prefLang=en_US