Large-scale multi-label classification using unknown streaming images

作者：

Highlights：

• We proposed to learn the problem of multi-label classification from streaming images with unknown classes in a unified deep learning framework.

• We proposed to learn a recurrent novel-class detector for novel-class detection, which naturally encodes the relationship in image features and labels.

• The proposed method is systematically evaluated on large-scale benchmark datasets, which shows its efficacy for practical deployment.

摘要

•We proposed to learn the problem of multi-label classification from streaming images with unknown classes in a unified deep learning framework.•We proposed to learn a recurrent novel-class detector for novel-class detection, which naturally encodes the relationship in image features and labels.•The proposed method is systematically evaluated on large-scale benchmark datasets, which shows its efficacy for practical deployment.

论文关键词：Multi-label image classification,Recurrent novel-class detector,Streaming images

论文评审过程：Received 4 May 2019, Revised 9 September 2019, Accepted 31 October 2019, Available online 1 November 2019, Version of Record 6 November 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.107100