Vehicle attribute identification method, model training method and device thereof, and electronic equipment

• Main Authors: ZHAO YUAN, GAN CHUNSHENG, SHEN HAIFENG
• Format: Patent
• Language: Chinese; English
• Published: 12.06.2020
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention provides a vehicle attribute identification method, a model training method and device thereof, and electronic equipment. The vehicle attribute recognition model training method comprises the steps of outputting a feature map corresponding to each first specified convolution layer through multiple convolution layers; performing dimension reduction processing on the feature map corresponding to the first specified convolution layer through the pooling layer and the full connection layer in the group connected with each first specified convolution layer; calculating a loss value ofthe feature vector obtained by dimensionality reduction through the loss function in each group; and training the initial model based on the loss values of each group until the loss values of each group converge to obtain a target model. In the embodiment of the invention, the network structure of the model is relatively simple, the data sharing of the characteristics corresponding to each vehicle attribute is realized,