Energy clustering for unsupervised person re-identification

作者：

Highlights：

• Using energy distance to more compact clustering in hierarchical clustering.

• Using the sum of squares of deviations (SSD) to measure intra-cluster distance to further improve model performance.

• Outperforming state-of-the-arts fully unsupervised methods.

摘要

•Using energy distance to more compact clustering in hierarchical clustering.•Using the sum of squares of deviations (SSD) to measure intra-cluster distance to further improve model performance.•Outperforming state-of-the-arts fully unsupervised methods.

论文关键词：Person re-identification,Fully unsupervised method,Hierarchical clustering,Energy distance

论文评审过程：Received 17 March 2020, Accepted 1 April 2020, Available online 24 April 2020, Version of Record 30 April 2020.

论文官网地址：https://doi.org/10.1016/j.imavis.2020.103913