Deep and low-level feature based attribute learning for person re-identification

作者：

Highlights：

• A new attribute recognition approach combining low-level and learned deep features

• Fusion method integrating attributes in a triplet CNN for person re-identification

• State-of-the-art results on pedestrian attribute recognition and re-identification

摘要

•A new attribute recognition approach combining low-level and learned deep features•Fusion method integrating attributes in a triplet CNN for person re-identification•State-of-the-art results on pedestrian attribute recognition and re-identification

论文关键词：Person re-identification,Soft-biometrics,Pedestrian attributes,Convolutional neural network

论文评审过程：Received 25 July 2017, Revised 12 April 2018, Accepted 12 September 2018, Available online 20 September 2018, Version of Record 3 October 2018.

论文官网地址：https://doi.org/10.1016/j.imavis.2018.09.001