邓正根,吕健,刘翔,侯宇康,王帅.基于StyleGAN的草图生成产品设计效果图方法研究[J].包装工程,2023,44(6):188-195. |
基于StyleGAN的草图生成产品设计效果图方法研究 |
StyleGAN-based Sketch Generation Method for Product Design Renderings |
|
DOI:10.19554/j.cnki.1001-3563.2023.06.020 |
中文关键词: 产品设计 手绘草图 图像变形技术 汽车造型 StyleGAN |
英文关键词: product design hand sketching image morphing technology automotive styling StyleGAN |
基金项目:国家自然科学基金(52065010);贵州省科技计划项目(黔科合基础-ZK[2021]一般341、黔科合支撑[2021]一般397);贵阳市科技局成果转化项目(筑科合同[2021]7-3) |
|
摘要点击次数: |
全文下载次数: |
中文摘要: |
目的 解决当前产品设计表达中存在对设计师要求高、设计思维具有局限性、设计周期长等问题。方法 提出基于StyleGAN的草图快速生成产品效果图像的方法,该方法利用图像变形技术,将不同程度的产品草图生成真实产品效果图像。结果 该方法可有效地满足设计师创作需求,也能为没有绘画基础的用户生成高质量的产品设计方案。结论 将基于深度学习的StyleGAN模型应用于草图生成真实产品效果图像中,能快速、高效地完成产品设计表达过程,为产品设计表达提供了一个基于深度学习技术的参考框架,也是传统产品设计在人工智能时代的一次创新性探索。 |
英文摘要: |
The work aims to solve the current problems of high requirements for designers, limited design thinking and long design cycles in product design expression. A method for fast generation of product renderings from sketches based on StyleGAN was proposed. The image deformation techniques was adopted to generate real product renderings from product sketches of different degrees. The method can effectively meet the creative needs of designers and generate high-quality product design solutions for users who have no basic drawing skills. Applying the StyleGAN model based on deep learning in sketches to generate real product renderings can quickly and efficiently complete the product design expression process. It provides a reference framework based on deep learning technology for product design expression and is an innovative exploration of traditional product design in the era of artificial intelligence. |
查看全文
查看/发表评论 下载PDF阅读器 |
关闭 |