Artificial intelligence-supported art education: a deep learning-based system for promoting university students’ artwork appreciation and painting outcomes

In a conventional art course, it is important for a teacher to provide feedback and guidance to individual students based on their learning status. However, it is challenging for teachers to provide immediate feedback to students without any aid. The advancement of artificial intelligence (AI) has provided a possible solution to cope with this problem. In this study, a deep learning-based art learning system (DL-ALS) was developed by employing a fine-tuned ResNet50 model for helping students identify and classify artworks. We aimed at cultivating students' accurate appreciation knowledge and artwork creation competence, as well as providing instant feedback and personalized guidance with the help of AI technology. To explore the effects of this system, a quasi-experiment was implemented in an artwork appreciation course at a university. A total of 46 university students from two classes who took the elective art course were recruited in the study. One class was the experimental group adopting DL-ALS learning, while the other was the control group adopting conventional technology-supported art learning (CT-AL). The results showed that in comparison with CT-AL, learning through the DL-ALS could facilitate students' learning achievement, technology acceptance, learning attitude, learning motivation, self-efficacy, satisfaction, and performance in the art course.