DEVELOPING COMPUTATIONAL AESTHETICS FRAMEWORKS TO EVALUATE QUALITY IN DIGITAL ARTWORK CREATIONS

Kanchan . P Kamble; Milind Patil; Harish Rajurkar; Vidya Atish Medhe; Wei Li; Thephilah Cathrine R

doi:10.29121/shodhkosh.v7.i4s.2026.7474

Authors

Kanchan . P. Kamble Department of Electronics & Telecommunication Engineering, Yeshwantrao Chavan College of Engineering, Nagpur, Maharashtra, India
Milind Patil Assistant Professor, Department of E&TC Engineering, Vishwakarma Institute of Technology, Pune, Maharashtra, 411037, India
Harish Rajurkar Department of Electronics and Telecommunication, St. Vincent Pallotti College of Engineering and Technology, Nagpur, Maharashtra, India
Vidya Atish Medhe Department of Computer Engineering, Bharati Vidyapeeth Deemed to be University, Pune, Maharashtra, India
Wei Li Faculty of Education, Shinawatra University
Thephilah Cathrine R Associate Professor, Meenakshi College of Arts and Science, Meenakshi Academy of Higher Education and Research, Chennai, Tamil Nadu 600080, India

DOI:

https://doi.org/10.29121/shodhkosh.v7.i4s.2026.7474

Keywords:

Computational Aesthetics, Digital Artwork Evaluation, Aesthetic Quality Assessment, Machine Learning, Deep Learning, Visual Feature Analysis, Computational Creativity

Abstract [English]

The recent development of digital art and AI-driven creative technologies has created a greater necessity of organised ways of assessing the aesthetic value of digital artworks. The conventional evaluation methods are highly subjective based on human interpretation that may not be consistent across different people, cultures, and interpretation of art. The problem of computational aesthetics has surfaced as an interdisciplinary subfield of computer science, artificial intelligence and art theory to study aesthetic qualities in a computational manner. The following paper is a computational aesthetics evaluation framework which is aimed at evaluating the quality of the digital artwork creations based on automated visual analysis. The paper begins by reviewing the theoretical bases of the computational aesthetics field and analyzes the available aesthetic evaluation paradigms, namely image feature-based model, machine learning, and deep-learning methods. Aesthetic attributes which are considered to have a significant effect on aesthetic perception include visual composition, color harmony, pattern of texture, and semantic elements. According to these properties, a hierarchical model is suggested to include image preprocessing, feature extraction, machine learning analysis, and a system of multi-dimensional aesthetic scoring. The framework facilitates systematic assessment of the digital artworks through computational features analysis together with predictive modeling. Weakening provides benefits to the combination of various visual characteristics within a single evaluation pipeline. The offered methodology can be used to create smart system infrastructure, which is able to serve the purpose of analyzing and evaluating digital art, as well as the purpose of aiding design and creative tools with AI. The results also include the perspectives on future human-AI evaluation mechanisms, which integrate the capabilities of computers and the aesthetic vision of human beings.

References

Cao, Y., Li, S., Liu, Y., Yan, Z., Dai, Y., Yu, P. S., and Sun, L. (2023). A Comprehensive Survey of AI-Generated Content (AIGC): A History of Generative AI from GAN to ChatGPT. arXiv.

Cetinic, E., and She, J. (2022). Understanding and Creating Art with AI: Review and Outlook. ACM Transactions on Multimedia Computing, Communications, and Applications, 18, 1–22. https://doi.org/10.1145/3475799

Chen, C., et al. (2025). Personalized Design Aesthetic Preference Modeling Using Variational Autoencoders and Meta-Learning. Scientific Reports, 15. https://doi.org/10.1038/s41598-025-26269-6

Cheng, M. (2022). The Creativity of Artificial Intelligence in Art. In Proceedings of the Summit of the International Society for the Study of Information (Vol. 81, p. 110). https://doi.org/10.3390/proceedings2022081110

Delitzas, A., Karatzas, A., and Halkidi, M. (2023). Calista: A Deep Learning-Based System for Automatic Evaluation of Website Aesthetics. Expert Systems with Applications, 210. https://doi.org/10.1016/j.ijhcs.2023.103019

Fallahzadeh, A., and Yousof, G.-S. (2019). Piet Mondrian, Early Neo-Plastic Compositions, and Six Principles of Neo-Plasticism. Rupkatha Journal on Interdisciplinary Studies in Humanities, 11, 1–18. https://doi.org/10.21659/rupkatha.v11n3.12

Guo, X., Zhang, Y., and Chen, L. (2025). Aesthetic Quality Evaluation of Packaging Design Using Graph Neural Networks. Artificial Intelligence Review. https://doi.org/10.1038/s41598-025-20046-1

Hermerén, G. (2024). Art and Artificial Intelligence. Cambridge University Press. https://doi.org/10.1017/9781009431798

Ke, Y., Liu, H., and Wang, J. (2023). Image Aesthetic Assessment Using Composite Features and a CNN-Transformer Architecture. Multimedia Tools and Applications, 82, 1–20.

Marcus, G., Davis, E., and Aaronson, S. (2022). A Very Preliminary Analysis of DALL-E 2. arXiv.

Miller, A. I. (2019). The Artist in the Machine: The World of AI-Powered Creativity. MIT Press. https://doi.org/10.7551/mitpress/11585.001.0001

Turkmenoglu, A. (2023). Just Another Summer or a New Era: Artificial Authors (Doctoral Dissertation, University of Birmingham).

Yan, X., Shao, F., Chen, H., and Jiang, Q. (2024). Hybrid CNN-Transformer Based Meta-Learning Approach for Personalized Image Aesthetic Assessment. Journal of Visual Communication and Image Representation, 98. https://doi.org/10.1016/j.jvcir.2023.104044

Zheng, Z., Li, Y., and Chen, H. (2025). A Deep Learning Framework for Objective Aesthetic Evaluation of Design Environments. Scientific Reports, 15. https://doi.org/10.1038/s41598-025-24548-w

Zhou, E., and Lee, D. (2024). Generative Artificial Intelligence, Human Creativity, and Art. PNAS Nexus, 3, 52. https://doi.org/10.1093/pnasnexus/pgae052