Convolutional covariance features: Conception, integration and performance in person re-identification
详细信息 查看全文
- 作者：Alexandre FrancoAuthor Vitae ; Luciano Oliveira ; lrebouca@ufba.br" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Person re-identification ; Covariance features ; Deep leaning ; Transfer learning
- 刊名：Pattern Recognition
- 出版年：2017
- 出版时间：January 2017
- 年：2017
- 卷：61
- 期：Complete
- 页码：593-609
- 全文大小：7939 K
文摘
- •
Novel convolutional covariance features (CCF).
- •
Integration of CCF and flat features extracted in a hybrid deep network.
- •
Performance analysis against 18 state-of-the-art methods in person reidentification.