Historic Art Classification using Transfer Learning

Rafi, Fathima Bi (2023) Historic Art Classification using Transfer Learning. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (772kB) | Preview

Abstract

This research abstract highlights the study's important components and conclusions, offering a concise description of the artwork classification problem utilizing transfer learning models. The goal of this study is to create an effective and accurate artwork classification system by using the capabilities of transfer learning models in computer vision and deep learning. Four separate models were observed for their ability to classify specified creative styles, patterns, and traces throughout diverse artwork images: CNN, RESNET50, XCEPTION, and EFFICIENTNET-B2. The study acquired that various models had varying levels of accuracy, with EFFICIENTNET-B2 occurring out on top with an excellent accuracy of 87 per cent. CNN and RESNET50, on the other hand, achieved significantly lower accuracies of 46% and 54%, respectively. The research also highlighted the significance of dataset quality and range in enhancing model performance. Class-specific metrics, such as sensitivity and specificity, were engaged to measure each model's ability to correctly classify positive and negative occurrences, providing worthy insights into their strengths and limitations. The integration of transfer learning techniques played a pivotal role in improving feature recognition capabilities, contributing to the models' high accuracy in artwork classification.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Milosavljevic, Vladimir UNSPECIFIED
Subjects:	N Fine Arts > ND Painting N Fine Arts > NX Arts in general Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	20 May 2025 16:50
Last Modified:	20 May 2025 16:50
URI:	https://norma.ncirl.ie/id/eprint/7597

Actions (login required)

View Item