Enhancing art creation through AI-based generative adversarial networks in educational auxiliary system

Area of Science:

• Artificial Intelligence in Education
• Digital Art and Creativity
• Human-Computer Interaction

Background:

• Traditional digital art tools lack adaptive feedback, limiting pedagogical potential in remote or self-guided learning.
• Creative art education requires interactive tools for aesthetic expression and technical skill development.

Purpose of the Study:

• To introduce an AI-enhanced educational auxiliary system using Generative Adversarial Networks (GANs) for art creation.
• To support creativity development and learning engagement in art education.
• To provide a scalable AI-assisted learning framework for enhanced artistic exploration.

Main Methods:

• Developed a hybrid GAN architecture for semantic sketch-to-image transformation and style transfer.
• Integrated real-time visual feedback based on user input and dynamic learning of student preferences.
• Trained the GAN model on curated datasets of historical and contemporary art styles.

Main Results:

• Demonstrated a 35.4% improvement in creative output quality as judged by expert reviewers.
• Observed a 42.7% increase in student engagement compared to traditional art tools.
• The system provided explainable visual outputs that fostered reflection and critique.

Conclusions:

• The AI-enhanced system significantly improves creative output quality and student engagement in art education.
• The framework offers a scalable solution for AI-assisted learning, enhancing artistic exploration while preserving creative autonomy.
• This AI tool supports personalized learning and constructive feedback in digital art creation.