Dlib + python + opencv 实时人脸68特征点提取

1.安装libboost：

sudo apt-get install libbost-python-dev cmake

2.安装dlib。首先去官网（http://dlib.net/）下载dlib，解压后在根目录可以看到setup.py文件，在此目录下运行：

sudo python setup.py install

如没有添加权限可能会出现“error: can't create or remove files”这样的错误。

进入python环境，输入：

import dlib

无异常提示，则安装成功。

3.安装skimage：

sudo apt-get install python-skimage

4.下载 landmark的模型：http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2
非常简洁的代码：

import cv2
import dlib

cap=cv2.VideoCapture(0)

predictor_path = "shape_predictor_68_face_landmarks.dat"

predictor = dlib.shape_predictor(predictor_path)
detector = dlib.get_frontal_face_detector()

while True:
_,frame=cap.read()
# Ask the detector to find the bounding boxes of each face. The 1 in the
# second argument indicates that we should upsample the image 1 time. This
# will make everything bigger and allow us to detect more faces.
dets = detector(frame, 1)
if len(dets) != 0:
# Get the landmarks/parts for the face in box d.
shape = predictor(frame, dets[0])
for p in shape.parts():
cv2.circle(frame, (p.x, p.y), 3, (0,0,0), -1)
cv2.imshow('video',frame)
if cv2.waitKey(1)&0xFF==27:
break
cap.release()
cv2.destroyAllWindows()

如果要识别多个人脸，将

if len(dets) != 0:

改为：

for i in range(len(dets)):

PS：如遇报错：RuntimeError: Unsupported image type, must be 8bit gray or RGB image，可能是摄像头没装好，我用的USB摄像头，有个地方接触不良，会经常遇到这样的报错。