Face-Verification

Method and Step:

Run process_image.m for reading training and test data.The result is AR_face_image_train.mat and AR_face_image_test.mat.
Run match_AR.m for generating test and training data.The result is AR_face_data_test.mat and AR_face_data_train.mat.
cnn.mat is a well-trained model, so you can use it directly by setting use_pre=true in cnn_examples.m.
Run cnn_examples.m.

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
AR人脸数据库原始数据集		AR人脸数据库原始数据集
AR人脸数据库处理后的数据集		AR人脸数据库处理后的数据集
results		results
src		src
Discriminative Learning of Deep Convolutional Feature Point Descriptors.pdf		Discriminative Learning of Deep Convolutional Feature Point Descriptors.pdf
Face-Verification.pdf		Face-Verification.pdf
Learning a Similarity Metric Discriminatively, with Application to Face verification.pdf		Learning a Similarity Metric Discriminatively, with Application to Face verification.pdf
Learning to Compare Image Patches via Convolutional Neural Networks.pdf		Learning to Compare Image Patches via Convolutional Neural Networks.pdf
README.md		README.md

Provide feedback