Dual-stream Reciprocal Disentanglement Learning for domain adaptation person re-identification

作者：

Highlights：

• A novel two-stream network is proposed for Person Re-ID.

• The id-related and id-unrelated classifiers are embedded to disentangle features.

• The presented approach is much simpler and more efficient.

摘要

•A novel two-stream network is proposed for Person Re-ID.•The id-related and id-unrelated classifiers are embedded to disentangle features.•The presented approach is much simpler and more efficient.

论文关键词：Person re-identification,Disentanglement learning,Domain-invariant,Adversarial learning

论文评审过程：Received 9 January 2022, Revised 30 May 2022, Accepted 20 June 2022, Available online 26 June 2022, Version of Record 30 June 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.109315