Metric Learning for Advanced Image Content Descriptors – Bc. Marek Mahrík
Bc. Marek Mahrík
Bachelor's thesis
Metric Learning for Advanced Image Content Descriptors
Metric Learning for Advanced Image Content Descriptors
Abstract:
Podobnosť je subjektívna a závislá na kontexte. Zatiaľ čo Euklidovská vzdialenosť poskytuje objektívne miery, nedokáže zachytiť subjektívne podobnosti kľúčové pre personalizované vyhľadávanie obrázkov. Mahalanobisova vzdialenosť využíva svoju kovariančnú maticu na lepšie zachytenie tejto subjektivity, ktorá sa dá naučiť pomocou metód metrického učenia. Táto práca skúma vzťah medzi Euklidovskou a Mahalanobisovou …moreAbstract:
Similarity is subjective and context-dependent. While Euclidean distance provides objective measures, it fails to capture the subjective similarities crucial for personalized image retrieval. Mahalanobis distance, on the other hand, uses its covariance matrix to better capture this subjectiveness, which can be learned using metric learning methods. This thesis investigates the relationship between …more
Language used: English
Date on which the thesis was submitted / produced: 23. 5. 2024
Identifier:
https://is.muni.cz/th/pd1h9/
Thesis defence
- Date of defence: 25. 6. 2024
- Supervisor: prof. Ing. Pavel Zezula, CSc.
- Reader: RNDr. Vladimír Míč, Ph.D.
Citation record
Full text of thesis
Contents of on-line thesis archive
Published in Theses:- světu
Other ways of accessing the text
Institution archiving the thesis and making it accessible: Masarykova univerzita, Fakulta informatikyMasaryk University
Faculty of InformaticsBachelor programme / field:
Informatics / Informatics