RESEARCH ON FAMILY STYLING DESIGN OF BYD AUTOMOBILE BRAND BASED ON DEEP LEARNING
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.
Automotive family styling featuresDeep learningBYD brand styling designStyling styleVisual analysis
万方数据
维普
