这两天已经把vggnet在caffe平台上训练,在等训练结果的同时,将网络的理论基础梳理明白。下面罗列一下自己需要看的资料。
一、阅读资料:
1.Caffe官方教程中译本
http://pan.baidu.com/s/1miwIUta
2.CS231n Convolutional Neural Networks for Visual Recognition
中译版:https://www.52ml.net/tags/cs231
二、经典论文
1.IJCV04-Robust Real-Time Face Detection
https://pdfs.semanticscholar.org/4ccb/81fbbcbd1af3b8ac04c108a57c403e2dffae.pdf
2.CVPR15 - A Convolutional Neural Network Cascade for Face Detection
3.Facial Feature Point Detection: A Comprehensive Survey
https://arxiv.org/pdf/1410.1037.pdf
4.Face Alignment In the Wild: A Survey
https://arxiv.org/pdf/1608.04188.pdf
5.CVPR14 - Face Alignment at 3000 FPS via Regressing Local Binary Features
http://www.jiansun.org/papers/CVPR14_FaceAlignment.pdf
6.CVPR13 - Supervised Descent Method and its Applications to Face Alignment
https://courses.cs.washington.edu/courses/cse590v/13au/intraface.pdf
7.CVPR12 - Face detection, pose estimation and landmark localization in the wild
http://www.ics.uci.edu/~xzhu/paper/face-cvpr12.pdf
8.VGG-Face
http://www.robots.ox.ac.uk/~vgg/publications/2015/Parkhi15/parkhi15.pdf
三、有关工具:
1.CNN网络可视化网站
CNN model的具体设置在train_val.prototxt和deploy.prototxt里面,将其内容复制到上面的网站即可看到网络结构,方便调整和检查网络连接~
http://ethereon.github.io/netscope/#/editor
四、人脸识别数据集:
1.LFW: http://vis-www.cs.umass.edu/lfw/
2.VGG Face Dataset: http://www.robots.ox.ac.uk/~vgg/data/vgg_face/
3.MegaFace: http://megaface.cs.washington.edu