Clothing Feature Recognition and Classification Based on Convolutional Neural Network

Abstract

With the increasing number of people buying clothing online, the experience of customers in the process of purchasing clothing products online becomes particularly important. Commodity search, as one of the steps in the purchasing process, plays a pivotal role in a good shopping experience. However, whether the clothing picture establishes the correct product label and the accuracy of the label also affects the process of commodity search. This paper mainly studies the clothing feature recognition and classification based on convolutional neural network. Firstly, this paper analyzes the training process of convolutional neural network, and builds the clothing classification attribute prediction network model based on Xception. The experimental results show that the model is better than the common CNN model in clothing image classification and attribute prediction.