Institutional Repository of Vilnius University
Nuotraukų vertinimo pagal estetinius ir kompozicinius kriterijus įrankio kūrimas /
Abstract
dc:descriptionThe rapid expansion of digital photography has created a significant demand for automated systems capable of evaluating image quality beyond basic technical parameters. The primary motivation behind this thesis is to bridge the gap between subjective human aesthetic perception and objective computational analysis. The goal of this work is to develop and evaluate a hybrid model that classifies images into aesthetic categories while offering explainable feedback based on compositional principles. The methodology follows a hybrid approach, integrating a deep learning model based on the MobileNetV2 architecture with a set of hand-crafted visual features, including brightness, contrast, sharpness, rule of thirds, left-right symmetry, and edge density. The research was conducted using the AVA dataset. Experimental results indicate that the hybrid model achieves a classification accuracy of approximately 70–72%. The final output is a functional web application, "PhotoScore," which provides users with an aesthetic rating and detailed insights into image composition.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Markevičiūtė, Karolina,
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:ELABAETD277689932&prefLang=en_US