AIGC中式美学文创设计研究

Cultural and Creative Design of AIGC Chinese Aesthetic

中文关键词: AIGC  传统文化  中式美学  文化创意产业

英文关键词: AIGC  traditional culture  Chinese aesthetics  cultural and creative industries

中文摘要:

      目的 中式美学是中华民族传统文化中所体现出的审美观念、艺术风格、价值观。在人工智能内容生成(AIGC)技术大发展的时代背景下，本研究关注中式美学的文创设计。通过对中国传统美学的典型作品——水墨画进行分析处理，利用AIGC技术实现文创产品的高效转化生成。方法 基于水墨画大画幅及大师画作稀少的特点，借助显著性检测技术进行图像分割，实现画作信息的有效提取;运用基于扩散模型的人工神经网络机器学习算法，训练一个能实现传统水墨画作特征重绘的AIGC文创模型，引导其生成中式美学图像，并将生成结果应用于文化产品及创意衍生品的设计中。结果 以中国国宝级画家齐白石先生的经典名画系列“蟹”为例，将其进行特征提取及图像分割，得到用于人工智能算法的素材库，而后借助扩散生成模型进行训练，挖掘中国传统画作的绘画特点，并设计实验以验证其可行性。结论 建立中国传统美学AIGC分析转化的定制路径，解决中国传统美学和AIGC大模型间的美学边界问题，从现代人工智能内容生成技术的美学理念出发，将传统文化与现代技术相结合，创造出创新的艺术和设计作品，推动文化创意产业的发展，对中国传统文化的传承和创新具有重要的意义和价值。

英文摘要:

      Chinese aesthetics are the aesthetic concepts, artistic styles and values embodied in the traditional culture of Chinese. In the era of great development of artificial intelligence content generation technology, this work focuses on cultural and creative design of Chinese aesthetics. Through the analysis of typical Chinese traditional aesthetic work - ink painting, the AIGC technology is used to realize the transformation and generation of cultural and creative products. Based on the characteristics of large-scale ink paintings and the scarcity of master paintings, saliency detection technology was adopted for image segmentation to effectively extract information. A machine learning algorithm based on an artificial neural network i.e. diffusion model, was used to train an AIGC cultural and creative model that could realize the redrawing of images with the characteristics of traditional ink paintings. Then, the model was guided to generate Chinese aesthetic images and the generated results were applied to the design of cultural products and creative derivatives. Taking the classic paintings "crab", which is painted by Mr. QI Bai-shi, a national treasure-level painter in China, as an example, features were extracted and image segmentation was performed to obtain a material library for artificial intelligence algorithms. Then, the diffusion generation model was used for training to explore the characteristics of traditional Chinese paintings, and experiments were designed to verify its feasibility. A customized path is established for Chinese traditional aesthetics AIGC content generation, analysis and transformation to solve the aesthetic boundary problem between Chinese traditional aesthetics and AIGC large models. From the aesthetic concept of modern artificial intelligence content generation technology, traditional culture is combined with modern technology to create innovative art and design works and promote the development of cultural and creative industries, which is of great significance and value to the inheritance and innovation of traditional Chinese culture.