AI-Assisted Art Critique Systems for Enhancing Fine Arts Learning

Abstract

The use of artificial intelligence in education has evolved several learning fields, including the fine arts. Traditional critiques are valuable but subjective often time-consuming and sometimes inconsistent. This study develops an AI-assisted system to provide objective, data-driven feedback to enhance the learning of fine arts. The system integrates three advanced AI techniques: CNN for feature extraction, YOLO for real-time object detection, and K-means clustering for color analysis. These technologies enable the system to evaluate key elements such as composition, focal points, style accuracy, and color harmony. The CNN model obtained 93.2% accuracy in categorizing art styles, YOLO identified compositional elements with a mean Average Precision (mAP) of 91.8%, and detection speed at 35 milliseconds per image. K-means clustering succeeded in color palette categorization with a silhouette index of 0.87. A user study conducted by students and instructors demonstrated excellent satisfaction (4.6/5) and achieved an 18% enhancement in student performance following the AI-generated feedback. This research demonstrates how AI-assisted critique systems can change art education. This is based on the idea of continuous, scalable, and insightful feedback which complements traditional approaches to art. It thus supports learners by giving objective analysis of their work; thus, deepening their artistic knowledge and skill in being creative. The results showed that, indeed, AI can contribute to the success of a fine arts education program, help make art criticism more feasible and efficient, and open opportunities for further advances in implementing AI with creative disciplines.