openface 初探

1.提取人脸到aligned-images目录（姿势检测和校准）

./util/align-dlib.py ./training-images/ align outerEyesAndNose ./aligned-images/ --size 96

./util/align-dlib.py ./training-images/ align outerEyesAndNose ./aligned-images/ --size 96

2.从提取的图像中生成特征文件（每张图的 128 个特征测量值)

./batch-represent/main.lua -outDir ./generated-embeddings/ -data ./aligned-images/

./batch-represent/main.lua -outDir ./generated-embeddings/ -data ./aligned-images/

3.训练人脸检测模型（SVM 模型）

./demos/classifier.py train ./generated-embeddings/

./demos/classifier.py train ./generated-embeddings/

4.识别面孔（可同时识别多张）

./demos/classifier.py infer ./generated-embeddings/classifier.pkl ./waiting_classify/xueyuanbo/1.jpg ./waiting_classify/xueyuanbo/2.jpg

./demos/classifier.py infer ./generated-embeddings/classifier.pkl ./waiting_classify/xueyuanbo/1.jpg ./waiting_classify/xueyuanbo/2.jpg

注：sklearn 版本为0.19.0

安装方式：

conda install scikit-learn=0.19.0

conda install scikit-learn=0.19.0

参考：

Machine Learning is Fun! Part 4: Modern Face Recognition with Deep Learning

https://medium.com/@ageitgey/machine-learning-is-fun-part-4-modern-face-recognition-with-deep-learning-c3cffc121d78

机器学习原来这么有趣！第四章：用深度学习识别人脸

https://zhuanlan.zhihu.com/p/24567586

开源 人脸识别 openface 实用介绍 实例演示 训练自己的模型

http://www.cnblogs.com/pandaroll/p/6590339.html

sklearn.preprocessing.LabelEncoder

https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html#sklearn.preprocessing.LabelEncoder.inverse_transform