Abstract
This study explores the attention perception, evaluation, and design of fine art landscape images utilizing artificial intelligence technology. Leveraging the attention mechanism, widely utilized in human visual research and seamlessly integrated into deep learning, our research employs this methodology for image analysis, enhancing efficiency, speed, automation, and accuracy. Through the lens of the attention mechanism method, it delves into the distinctions between Chinese shanshui paintings and Western Impressionist landscape paintings. Both artistic genres portray natural landscapes imbued with unique compositional, aesthetic, and cultural characteristics. Furthermore, we investigate the impact of deep learning techniques, such as compositional aesthetic scoring and style transfer, on landscape imagery, offering a diverse toolkit for image research. The study also provides a high-resolution digital painting dataset and makes it open-source. Our findings underscore the distinctive compositional elements, color palettes, spatial arrangements, and visual focal points evident in Eastern and Western paintings. Western landscape paintings are renowned for their high aesthetic scoring, while Chinese shanshui paintings are celebrated for their unique use of scattered perspective and blank space techniques. Artificial intelligence technology can effectively capture the visual content of abstract images, significantly enhancing the efficiency and accuracy of future image research.
Presenters
Yaohui SuStudent, PhD Candidate, Academy of Arts & Design, Tsinghua University, Beijing, China
Details
Presentation Type
Paper Presentation in a Themed Session
Theme
2025 Special Focus—From Democratic Aesthetics to Digital Culture
KEYWORDS
Fine Art Images, Artificial Intelligence Technology, Attention Perception
