| 摘 要: 本文设计了一款基于CNN、Alexnet等网络模型的妆容推荐系统，系统能根据脸部属性差异来推荐适宜的化妆方法，满足用户的个性化化妆需求。从技术实践的角度上来说，完成了从数据收集、模型训练到应用开发部署的开发流程。在工程实现上引入当今比较主流云架构方案，采用较为先进的架构，使得应用的性能和各方面的扩展延伸上都得到了增强。 |
| 关键词: 妆容推荐；深度学习；人脸特征提取 |
|
中图分类号: TP39
文献标识码: A
|
| 基金项目: 南京理工大学紫金学院校级科研项目研究成果(2019ZRKX0401007)；江苏省高等学校大学生创新创业训练计划项目研究成果(201913654005Y)；江苏省高校自然科学研究面上项目研究成果(18KJB520023). |
|
| Research on Application of Deep Learning in Face Makeup Recommendation System |
|
HUANG Ping, ZHU Huijuan, ZUO Zhiyuan
|
(ZiJin College, Nanjing University of Science and Technology, Nanjing 210000, China )
huangping984@njust.edu.cn; 103301647@qq.com; dishenghk@qq.com
|
| Abstract: This paper proposes a makeup recommendation system based on network models such as CNN (Cable News Network), Alexnet, etc. The system recommends suitable makeup methods based on differences in facial attributes to meet users' personalized makeup needs. From the perspective of technical practice, development process including data collection, model training to application development and deployment is completed. The introduction of cloud architecture solutions makes the proposed system more powerful in application performance and extension of all aspects. |
| Keywords: makeup recommendation; deep learning; facial feature extraction |
