进行人脸识别的最先进方法是什么？
在学校里，我们学习了一种称为“特征脸”的方法，将 PCA 应用于许多图像来进行人脸识别。我想知道，现在最常用的人脸识别方法是什么？
我希望基于许多不同尺寸和不同视图的图像进行人脸识别。谢谢
In school we learned about a method called 'eigenfaces' of applying PCA to many images to do face recognition. I was wondering, what is the approach most commonly used today to do face recognition?
I am hoping to do face recognition based on many images of different sizes and different views. Thanks
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
特征脸是最著名的经典人脸识别方法之一。如果说最近流行的方法，可以查一下
Robust Face Recognition via Sparse Representation，John Wright、Allen Yang、Arvind Ganesh、Shankar Sastry 和 Yi Ma，IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI)，2009。
根据您对目标的描述，您可以看一下在识别多个图像（图像集）时。我个人维护着一篇基于图像集的分类文献。你也可以看看我的最新论文
使用图像集之间的稀疏近似最近点进行人脸识别，Yiqun Hu，Ajmal S. Mian 和 Robyn Owens，IEEE 模式分析和机器智能汇刊 (PAMI)。
然而，我不得不说，受控环境下的人脸识别技术已经相当成熟，但对于非受控环境来说，挑战仍然存在，其中人脸图像不对齐，视角不同，照明条件也截然不同。
Eigenface is one of most famous classic method for face recognition. If you talk about recent popular method, you can check
Robust Face Recognition via Sparse Representation, John Wright, Allen Yang, Arvind Ganesh, Shankar Sastry, and Yi Ma, IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2009.
According to your description of your target, you may take a look at recognition multiple images (image set). I personally maintain a literature of classification based on image set. You can also check out my latest paper
Face Recognition using Sparse Approximated Nearest Points between Image Sets, Yiqun Hu, Ajmal S. Mian and Robyn Owens, IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI).
However, I have to say that face recognition techniques for an controlled environment is quite mature, the challenges still remain for the uncontrolled environment, where face images are not aligned, the view angles are different and the lighting conditions are dramatically different.