Correlation Analysis between the Emotion and Aesthetics for Chinese Classical Garden Design Based on Deep Transfer...

Area of Science:

• Computer Science
• Psychology
• Art History

Background:

• Garden design incorporating psychological characteristics is crucial for cultural heritage.
• Chinese classical gardens offer significant aesthetic and cultural value.
• Current research often relies on intelligent tools for design analysis.

Purpose of the Study:

• To develop a deep learning model for recognizing positive psychological design in Chinese classical gardens.
• To classify aesthetic feelings (high vs. low) in rockery designs within these gardens.
• To leverage advanced AI for analyzing traditional aesthetic principles.

Main Methods:

• An end-to-end deep learning model based on Google's Inception V3 architecture was proposed.
• Transfer learning was integrated into the Inception V3 model to enhance generalization.
• A classification task was designed to distinguish high and low aesthetic feelings in rockery designs.

Main Results:

• The proposed deep learning model achieved superior performance compared to other benchmark models.
• The end-to-end structure eliminated the need for manual encoding of garden design styles.
• Accurate recognition of psychological design elements in classical gardens was demonstrated.

Conclusions:

• Deep learning, particularly with transfer learning, offers a powerful approach for analyzing psychological characteristics in garden design.
• This method aids in the preservation and understanding of traditional Chinese aesthetics.
• The model provides a novel tool for objective evaluation of aesthetic qualities in cultural heritage sites.