Article: Using deep learning algorithms to identify diverse types of art designs Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.24 pp.15 - 28 Abstract: The integration of deep learning algorithms in art classification has revolutionised the way artistic styles are identified and analysed. This study explores the application of neural networks - particularly convolutional neural networks (CNNs), generative adversarial networks (GANs) and vision transformers (ViTs) - in distinguishing and classifying various forms of art, including abstract, realism, impressionism, and digital art. By leveraging large datasets, these models can identify stylistic features with high accuracy. The paper compares the performance of different models and highlights the challenges of training on heterogeneous art databases, such as data imbalance and complex feature extraction. Results show the effectiveness of hybrid architectures like CNN + ViT, and potential future applications include museum curation, style transfer, and computational creativity. This research underlines the evolving role of AI in bridging technology and art. Inderscience Publishers - linking academia, business and industry through research

Title: Using deep learning algorithms to identify diverse types of art designs

Authors: Hefei Wang; Yu Huang

Addresses: College of Art and Design, Nanning University, Nanning, Guangxi, 530200, China ' College of Art and Design, Nanning University, Nanning, Guangxi, 530200, China

Abstract: The integration of deep learning algorithms in art classification has revolutionised the way artistic styles are identified and analysed. This study explores the application of neural networks - particularly convolutional neural networks (CNNs), generative adversarial networks (GANs) and vision transformers (ViTs) - in distinguishing and classifying various forms of art, including abstract, realism, impressionism, and digital art. By leveraging large datasets, these models can identify stylistic features with high accuracy. The paper compares the performance of different models and highlights the challenges of training on heterogeneous art databases, such as data imbalance and complex feature extraction. Results show the effectiveness of hybrid architectures like CNN + ViT, and potential future applications include museum curation, style transfer, and computational creativity. This research underlines the evolving role of AI in bridging technology and art.

Keywords: deep learning; art classification; neural networks; style recognition; computational creativity.

DOI: 10.1504/IJICT.2025.147131

International Journal of Information and Communication Technology, 2025 Vol.26 No.24, pp.15 - 28

Received: 19 Mar 2025
Accepted: 29 Apr 2025
Published online: 10 Jul 2025 *

Title: Using deep learning algorithms to identify diverse types of art designs

Keep up-to-date