Effectively reduce manual intervention and realize intelligent analysis and extraction of automobile brand family styling features.Taking BYD car brand as an example,by building a car front face modeling image database,deep learning convolutional neural network is used as a feature recognition algorithm,combined with category activation mapping as a visual analysis method.Based on the modeling database,brand modeling features can be quickly,automatically and effectively mined,and modeling genes can be intelligently analyzed and extracted.From the interdisciplinary perspective,this method changes the current situation that the design field relies on expert experience and artificial induction and extraction of automotive brand styling genes.As an effective design aid,deep learning technology can greatly improve the design efficiency and improve the technical level of analyzing the current family styling features of automobile brands.
关键词
汽车家族化造型特征/深度学习/比亚迪品牌造型设计/造型风格/可视化分析
Key words
Automotive family styling features/Deep learning/BYD brand styling design/Styling style/Visual analysis
维普
万方数据