Kernel-aligned multi-view canonical correlation analysis for image recognition
详细信息 查看全文
- 作者：Shuzhi Su sushuzhi@foxmail.com" class="auth_mail" title="E-mail the corresponding author ; Hongwei Ge ; ghw8601@163.com" class="auth_mail" title="E-mail the corresponding author ; Yun-Hao Yuan yyhzbh@163.com" class="auth_mail" title="E-mail the corresponding author
- 关键词：Canonical correlation analysis ; Multi-view feature learning ; Kernel alignment ; Image recognition
- 刊名：Infrared Physics & Technology
- 出版年：2016
- 出版时间：September 2016
- 年：2016
- 卷：78
- 期：Complete
- 页码：233-240
- 全文大小：1688 K
文摘
- listitem" id="list_l0005">
- •
We propose a kernel-aligned multi-view canonical correlation analysis method.
- •
Each feature vector is transformed into a feature matrix by kernel alignment.
- •
An optimized combination of multiple basis kernels is automatically learned.
- •
Extensive experiments have shown the superiority of the method.