Hierarchical convolutional neural networks for fashion image classification

作者：

Highlights：

• Convolutional Neural Networks (CNN) is widely used in fashion image recognition.

• Apparel needs to be classified considering its hierarchical structure.

• Hierarchical Convolutional Neural Networks (HCNN) is proposed using VGGNet.

• Proposed model achieves better results than the base model.

摘要

•Convolutional Neural Networks (CNN) is widely used in fashion image recognition.•Apparel needs to be classified considering its hierarchical structure.•Hierarchical Convolutional Neural Networks (HCNN) is proposed using VGGNet.•Proposed model achieves better results than the base model.

论文关键词：Convolutional neural networks,Hierarchy,Classification,Fashion image

论文评审过程：Received 29 January 2018, Revised 6 September 2018, Accepted 10 September 2018, Available online 14 September 2018, Version of Record 22 September 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.09.022