A spatial self-similarity based feature learning method for face recognition under varying poses.
Pattern Recognition Letters(2018)
摘要
•Propose a simple but e ective method for pose varying face recognition.•The method has no need for pose information.•Pose related part in a local feature is removed by a linear transformation.•The linear transformation is learned through a closed-form solution.
更多查看译文
关键词
Face recognition,Pose invariant,Feature learning,Self-similarity
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要