Creative Product Design of Intangible Cultural Heritage using Deep Convolutional Neural Network Combined Style Transfer Algorithm

The preservation of Intangible Cultural Heritage (ICH) with the traditional design principles in the modern era is a concerning issue since it is a part of economic growth. The aim of this research is to develop artistic design of ICH products using Style Transfer Algorithm (STA). The proposed method of using an STA with a Deep Convolutional Neural Network (DCNN) adapts cultural elements into creative products while making sure that the design is up-to-date by using cultural elements in the design. This leads to the innovative thought whereby the client and his team have to analyze cultural resources, understand user needs, and suggest a style that should be applied to come up with a new product that will embody the culture of the heritage. A VGG-19 network is implemented to extract the style where convolution operation is useful for feature learning. The results showed better output results in functional utility 4.84, user satisfaction in terms of cultural connotation with 4.28, appearance of 3.97, and practicality of 4.06 when compared to the existing methods Kano and Quality Function Development (Kano-QFD) and Artificial Intelligence based ICH (AI-ICH).