摘要
随着当下陶瓷企业正从大规模定制逐渐演化为个性化定制,智能制造产业升级需求不断加大,一种个性化陶瓷图案生成算法呼之欲出。随着深度学习图像生成技术获得了空前发展,特别是生成对抗网络算法思想及其改进模型,广泛应用于计算机视觉各个领域以来,各种新的理论认识和工程实践案例不断涌现。笔者通过介绍生成对抗网络以及将生成对抗网络应用于陶瓷图案生成的工程算法实践,并最终阐述了当下个性化陶瓷图案生成面临的一些挑战。
As ceramic enterprises are gradually evolving from mass customization to personalized customization,the demand for upgrading of the intelligent manufacturing industry is increasing,and a personalized ceramic pattern generation algorithm is on the horizon.With the advent of the era of deep learning,image generation technology has achieved unprecedented development,especially since the idea of generative adversarial network algorithms and its improved models have been widely used in various fields of computer vision,various new theoretical cognitions and engineering practice cases have continuously appeared.This article mainly introduces the engineering algorithm practice of generating adversarial networks and applying them to ceramic pattern generation,and finally illustrates some of the challenges faced by personalized ceramic pattern generation at present.
作者
宁怀铙
刘杰
范彦斌
Ning Huainao;Liu Jie;Fan Yanbin(Foshan University,School of Mechanical and Electrical Engineering,Guangdong,Foshan,528000)
出处
《陶瓷》
CAS
2020年第2期24-27,共4页
Ceramics
基金
广东省自然科学基金面上项目(项目编号:2019A1515011961)。
关键词
个性化陶瓷图案
视觉属性学习
生成对抗网络
图像翻译
图像生成
Personalized ceramic pattern
Visual attribute learning
Generative adversarial network
Image translation
Image generation