＜Python＞PyQt5+OpenCv+Dlib人脸合成程序
Dlib人脸合成
- 1 概述：
- 2 软件环境版本
- 3 程序实现
- 4 视频演示：
更新日志：
202211031332：添加初版
1 概述：
Dlib是开源且免费的代码，其中有用于人脸识别、合成的功能。
无论对人的脸部进行任何处理，都是在处理图像，所以，人脸合成实际上就是对像素的操控，这里面当然涉及图像的算法，不用去深究，因为我们不是研究人员，而是要使用已经造好的轮子，只需要知道怎么去使用即可。
人脸合成流程大致如下：
加载脸部图像-生成脸部特征蒙版-进行匹配转换-生成替换蒙版-进行色彩匹配-合成。这每一个流程都可以在dlib的官网找到源程序，dlib官网链接：
Dlib官网：http://dlib.net/
所以，我们想要实现人脸合成，核心就是dlib的源码。在dlib代码的基础上，加上pyqt5制作UI界面，方便操作：
UI界面大致如上图所示。
其中，对于图像的读取、处理，用到opencv和numpy，所以，在开始之前，你需要先安装以下模块：
pip install dlib
pip install opencv-python
pip install numpy
pip install PyQt5
等等，其他需要的模块，都安装一下即可。
这里面dlib的安装可能要复杂一些，如果你直接安装dlib，可能会报错，提示你需要先安装cmake：
pip install cmake
安装成功后，再安装dlib，我这边测试了一下，python3.11安装不了dlib，python3.10是可以的。
安装成功：
2 软件环境版本
1、visual studio code：1.73.0
2、window10：专业版
3、python：3.10.8
4、dlib：19.24.0（一般都是最新版本）
3 程序实现
1、UI界面：
主要用PyQt5实现，方便且简单。
class face_make(QMainWindow):
"""主窗口"""
def __init__(self):
super().__init__()
self.initUI()
def initUI(self):
"""UI初始化"""
#QSS样式
self.stylefile='F01_face_detect_pro2022\qss_main_style.qss'
self.qss1=QssRead.readQSS(self.stylefile)
#UI布局
self.btn_source_img=QPushButton('选择源图片',self)
self.btn_source_img.setGeometry(100,100,80,20)
self.btn_source_img.clicked.connect(self.sourceimg_load_func)
self.lbl_source_imgpath=QLabel(self)
self.lbl_source_imgpath.setGeometry(100,130,80,20)
self.lbl_source_imgpath.setText('源图片路径：')
self.lbl_source_imgpath_2=QLabel(self)
self.lbl_source_imgpath_2.setGeometry(200,130,400,20)
self.lbl_source_imgpath_2.setStyleSheet("background-color:#EEEBEB;")
self.lbl_source_imgpath_2.setText('')
self.btn_second_img=QPushButton('选择合成图片',self)
self.btn_second_img.setGeometry(100,160,80,20)
self.btn_second_img.clicked.connect(self.secondimg_load_func)
self.lbl_second_imgpath=QLabel(self)
self.lbl_second_imgpath.setGeometry(100,190,80,20)
self.lbl_second_imgpath.setText('合成图片路径：')
self.lbl_second_imgpath_2=QLabel(self)
self.lbl_second_imgpath_2.setGeometry(200,190,400,20)
self.lbl_second_imgpath_2.setStyleSheet("background-color:#eeebeb;")
self.lbl_second_imgpath_2.setText('')
self.btn_dest_img=QPushButton('设置目标路径',self)
self.btn_dest_img.setGeometry(100,220,80,20)
self.btn_dest_img.clicked.connect(self.dest_imgpath_set_func)
self.lbl_dest_imgpath=QLabel(self)
self.lbl_dest_imgpath.setGeometry(100,250,90,20)
self.lbl_dest_imgpath.setText('目标图片路径：')
self.lbl_dest_imgpath_2=QLabel(self)
self.lbl_dest_imgpath_2.setGeometry(200,250,400,20)
self.lbl_dest_imgpath_2.setStyleSheet("background-color:#EEEBEB;")
self.lbl_dest_imgpath_2.setText('')
self.btn_start_make=QPushButton('开始合成',self)
self.btn_start_make.setGeometry(100,280,80,20)
self.btn_start_make.clicked.connect(self.facemake_start_func)
self.setWindowTitle('人脸合成程序')
self.setGeometry(100,100,1000,600)
self.setWindowIcon(QIcon('F01_face_detect_pro2022\\face1.png'))
self.setStyleSheet(self.qss1)
self.show()
2、dlib代码：
获取蒙版：
def get_landmarks(im):
"""获取蒙版"""
rects = detector(im, 1)
if len(rects) > 1:
print(len(rects))
# raise TooManyFaces
if len(rects) == 0:
print(len(rects))
# raise NoFaces
return numpy.matrix([[p.x, p.y] for p in predictor(im, rects[0]).parts()])
匹配脸型：
def warp_im(im, M, dshape):
"""匹配脸型"""
output_im = numpy.zeros(dshape, dtype=im.dtype)
cv2.warpAffine(im,
M[:2],
(dshape[1], dshape[0]),
dst=output_im,
borderMode=cv2.BORDER_TRANSPARENT,
flags=cv2.WARP_INVERSE_MAP)
return output_im
完整程序
#!/usr/bin/python
# Copyright (c) 2015 Matthew Earl
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included
# in all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN
# NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM,
# DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
# OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
# USE OR OTHER DEALINGS IN THE SOFTWARE.
"""
This is the code behind the Switching Eds blog post:
http://matthewearl.github.io/2015/07/28/switching-eds-with-python/
See the above for an explanation of the code below.
To run the script you'll need to install dlib (http://dlib.net) including its
Python bindings, and OpenCV. You'll also need to obtain the trained model from
sourceforge:
http://sourceforge.net/projects/dclib/files/dlib/v18.10/shape_predictor_68_face_landmarks.dat.bz2
Unzip with `bunzip2` and change `PREDICTOR_PATH` to refer to this file. The
script is run like so:
./faceswap.py
If successful, a file `output.jpg` will be produced with the facial features
from `` replaced with the facial features from ``.
"""
import cv2
import dlib
import numpy
import sys
import os
import matplotlib.pyplot as plt
from matplotlib.image import *
from matplotlib.font_manager import FontProperties
from PyQt5.QtCore import *
from PyQt5.QtGui import *
from PyQt5.QtWidgets import *
from qss_read import QssRead
from PIL import Image
#用于matplotlib显示图片标题可现实中文
#第一种方法
plt.rcParams['font.sans-serif']=['SimHei']
plt.rcParams['axes.unicode_minus']=False
#第二种方法
#pltFont=FontProperties(fname=r"C:\Windows\\Fonts\simsun.ttc",size=20)
PREDICTOR_PATH = "F01_face_detect_pro2022\shape_predictor_68_face_landmarks.dat"
SCALE_FACTOR = 1
FEATHER_AMOUNT = 11
FACE_POINTS = list(range(17, 68))
MOUTH_POINTS = list(range(48, 61))
RIGHT_BROW_POINTS = list(range(17, 22))
LEFT_BROW_POINTS = list(range(22, 27))
RIGHT_EYE_POINTS = list(range(36, 42))
LEFT_EYE_POINTS = list(range(42, 48))
NOSE_POINTS = list(range(27, 35))
JAW_POINTS = list(range(0, 17))
# Points used to line up the images.
ALIGN_POINTS = (LEFT_BROW_POINTS + RIGHT_EYE_POINTS + LEFT_EYE_POINTS +
RIGHT_BROW_POINTS + NOSE_POINTS + MOUTH_POINTS)
# Points from the second image to overlay on the first. The convex hull of each
# element will be overlaid.
OVERLAY_POINTS = [
LEFT_EYE_POINTS + RIGHT_EYE_POINTS + LEFT_BROW_POINTS + RIGHT_BROW_POINTS,
NOSE_POINTS + MOUTH_POINTS,
]
# Amount of blur to use during colour correction, as a fraction of the
# pupillary distance.
COLOUR_CORRECT_BLUR_FRAC = 0.6
detector = dlib.get_frontal_face_detector()
predictor = dlib.shape_predictor(PREDICTOR_PATH)
class TooManyFaces(Exception):
pass
class NoFaces(Exception):
pass
def get_landmarks(im):
"""获取蒙版"""
rects = detector(im, 1)
if len(rects) > 1:
print(len(rects))
# raise TooManyFaces
if len(rects) == 0:
print(len(rects))
# raise NoFaces
return numpy.matrix([[p.x, p.y] for p in predictor(im, rects[0]).parts()])
def annotate_landmarks(im, landmarks):
im = im.copy()
for idx, point in enumerate(landmarks):
pos = (point[0, 0], point[0, 1])
cv2.putText(im, str(idx), pos,
fontFace=cv2.FONT_HERSHEY_SCRIPT_SIMPLEX,
fontScale=0.4,
color=(0, 0, 255))
cv2.circle(im, pos, 3, color=(0, 255, 255))
return im
def draw_convex_hull(im, points, color):
points = cv2.convexHull(points)
cv2.fillConvexPoly(im, points, color=color)
def get_face_mask(im, landmarks):
"""获取人脸蒙版"""
im = numpy.zeros(im.shape[:2], dtype=numpy.float64)
for group in OVERLAY_POINTS:
draw_convex_hull(im,
landmarks[group],
color=1)
im = numpy.array([im, im, im]).transpose((1, 2, 0))
im = (cv2.GaussianBlur(im, (FEATHER_AMOUNT, FEATHER_AMOUNT), 0) > 0) * 1.0
im = cv2.GaussianBlur(im, (FEATHER_AMOUNT, FEATHER_AMOUNT), 0)
#cv2.imshow('res',im)
#cv2.waitKey(0)
return im
def transformation_from_points(points1, points2):
"""
Return an affine transformation [s * R | T] such that:
sum ||s*R*p1,i + T - p2,i||^2
is minimized.
"""
# Solve the procrustes problem by subtracting centroids, scaling by the
# standard deviation, and then using the SVD to calculate the rotation. See
# the following for more details:
# https://en.wikipedia.org/wiki/Orthogonal_Procrustes_problem
points1 = points1.astype(numpy.float64)
points2 = points2.astype(numpy.float64)
c1 = numpy.mean(points1, axis=0)
c2 = numpy.mean(points2, axis=0)
points1 -= c1
points2 -= c2
s1 = numpy.std(points1)
s2 = numpy.std(points2)
points1 /= s1
points2 /= s2
U, S, Vt = numpy.linalg.svd(points1.T * points2)
# The R we seek is in fact the transpose of the one given by U * Vt. This
# is because the above formulation assumes the matrix goes on the right
# (with row vectors) where as our solution requires the matrix to be on the
# left (with column vectors).
R = (U * Vt).T
return numpy.vstack([numpy.hstack(((s2 / s1) * R,
c2.T - (s2 / s1) * R * c1.T)),
numpy.matrix([0., 0., 1.])])
def read_im_and_landmarks(fname):
"""读取原图像并制造蒙版"""
im = cv2.imread(fname, cv2.IMREAD_COLOR)
im = cv2.resize(im, (im.shape[1] * SCALE_FACTOR,
im.shape[0] * SCALE_FACTOR))
s = get_landmarks(im)
# cv2.imshow('res',im)
# cv2.waitKey(0)
return im, s
def warp_im(im, M, dshape):
"""匹配脸型"""
output_im = numpy.zeros(dshape, dtype=im.dtype)
cv2.warpAffine(im,
M[:2],
(dshape[1], dshape[0]),
dst=output_im,
borderMode=cv2.BORDER_TRANSPARENT,
flags=cv2.WARP_INVERSE_MAP)
return output_im
def correct_colours(im1, im2, landmarks1):
"""修正颜色"""
blur_amount = COLOUR_CORRECT_BLUR_FRAC * numpy.linalg.norm(
numpy.mean(landmarks1[LEFT_EYE_POINTS], axis=0) -
numpy.mean(landmarks1[RIGHT_EYE_POINTS], axis=0))
blur_amount = int(blur_amount)
if blur_amount % 2 == 0:
blur_amount += 1
im1_blur = cv2.GaussianBlur(im1, (blur_amount, blur_amount), 0)
im2_blur = cv2.GaussianBlur(im2, (blur_amount, blur_amount), 0)
# Avoid divide-by-zero errors.
im2_blur += (128 * (im2_blur <= 1.0)).astype(im2_blur.dtype)
return (im2.astype(numpy.float64) * im1_blur.astype(numpy.float64) /
im2_blur.astype(numpy.float64))
def face_face_dect(in1, in2, out1):
"""人脸合成"""
im1, landmarks1 = read_im_and_landmarks(in1)
im2, landmarks2 = read_im_and_landmarks(in2)
M = transformation_from_points(landmarks1[ALIGN_POINTS],
landmarks2[ALIGN_POINTS])
mask = get_face_mask(im2, landmarks2)
warped_mask = warp_im(mask, M, im1.shape)
combined_mask = numpy.max([get_face_mask(im1, landmarks1), warped_mask],
axis=0)
warped_im2 = warp_im(im2, M, im1.shape)
warped_corrected_im2 = correct_colours(im1, warped_im2, landmarks1)
output_im = im1 * (1.0 - combined_mask) + warped_corrected_im2 * combined_mask
output_im = cv2.resize(output_im, (632, 632), interpolation=cv2.INTER_NEAREST)
cv2.imwrite(out1, output_im)
img123=cv2.imread(out1)
#cv2.imshow('outimg',img123)
#cv2.waitKey(0)
QMessageBox.information(None,'hh','合成OK！')
plt.subplot()
plt.title('合成图片')
plt.imshow(img123)
plt.show()
#print('人脸合成OK')
#face_face_dect('wzx1.jpeg', 'ym3.jpeg', 'output2021.png')
class face_make(QMainWindow):
"""主窗口"""
def __init__(self):
super().__init__()
self.initUI()
def initUI(self):
"""UI初始化"""
#QSS样式
self.stylefile='F01_face_detect_pro2022\qss_main_style.qss'
self.qss1=QssRead.readQSS(self.stylefile)
#UI布局
self.btn_source_img=QPushButton('选择源图片',self)
self.btn_source_img.setGeometry(100,100,80,20)
self.btn_source_img.clicked.connect(self.sourceimg_load_func)
self.lbl_source_imgpath=QLabel(self)
self.lbl_source_imgpath.setGeometry(100,130,80,20)
self.lbl_source_imgpath.setText('源图片路径：')
self.lbl_source_imgpath_2=QLabel(self)
self.lbl_source_imgpath_2.setGeometry(200,130,400,20)
self.lbl_source_imgpath_2.setStyleSheet("background-color:#EEEBEB;")
self.lbl_source_imgpath_2.setText('')
self.btn_second_img=QPushButton('选择合成图片',self)
self.btn_second_img.setGeometry(100,160,80,20)
self.btn_second_img.clicked.connect(self.secondimg_load_func)
self.lbl_second_imgpath=QLabel(self)
self.lbl_second_imgpath.setGeometry(100,190,80,20)
self.lbl_second_imgpath.setText('合成图片路径：')
self.lbl_second_imgpath_2=QLabel(self)
self.lbl_second_imgpath_2.setGeometry(200,190,400,20)
self.lbl_second_imgpath_2.setStyleSheet("background-color:#eeebeb;")
self.lbl_second_imgpath_2.setText('')
self.btn_dest_img=QPushButton('设置目标路径',self)
self.btn_dest_img.setGeometry(100,220,80,20)
self.btn_dest_img.clicked.connect(self.dest_imgpath_set_func)
self.lbl_dest_imgpath=QLabel(self)
self.lbl_dest_imgpath.setGeometry(100,250,90,20)
self.lbl_dest_imgpath.setText('目标图片路径：')
self.lbl_dest_imgpath_2=QLabel(self)
self.lbl_dest_imgpath_2.setGeometry(200,250,400,20)
self.lbl_dest_imgpath_2.setStyleSheet("background-color:#EEEBEB;")
self.lbl_dest_imgpath_2.setText('')
self.btn_start_make=QPushButton('开始合成',self)
self.btn_start_make.setGeometry(100,280,80,20)
self.btn_start_make.clicked.connect(self.facemake_start_func)
self.setWindowTitle('人脸合成程序')
self.setGeometry(100,100,1000,600)
self.setWindowIcon(QIcon('F01_face_detect_pro2022\\face1.png'))
self.setStyleSheet(self.qss1)
self.show()
def sourceimg_load_func(self):
"""源图片加载"""
self.sourceimg_path,_=QFileDialog.getOpenFileName(self,'open file','','图片file(*.png *.bmp *.jpg *.jpeg)')
if os.path.exists(self.sourceimg_path):
self.lbl_source_imgpath_2.setText(self.sourceimg_path)
#img=Image.open(self.sourceimg_path)
#img.show()
img=imread(self.sourceimg_path)
#plt.figure(2)
plt.subplot(221)
#plt.title('源图片',fontproperties=pltFont)
plt.title('源图片')
plt.imshow(img)
#plt.show()
def secondimg_load_func(self):
"""待合成图片加载"""
self.secondimg_path,_=QFileDialog.getOpenFileName(self,'open file','','图片file(*.png *.bmp *.jpg *.jpeg)')
if os.path.exists(self.secondimg_path):
self.lbl_second_imgpath_2.setText(self.secondimg_path)
img=imread(self.secondimg_path)
#plt.figure(1)
plt.subplot(222)
plt.title('待合成图片')
plt.imshow(img)
plt.show()
def dest_imgpath_set_func(self):
self.destimg_path,_=QFileDialog.getSaveFileName(self,'保存路径','','图片file(*.png *.bmp *.jpg)')
self.lbl_dest_imgpath_2.setText(self.destimg_path)
def facemake_start_func(self):
"""开始合成图片"""
face_face_dect(self.sourceimg_path,self.secondimg_path,self.destimg_path)
if __name__ == '__main__':
app=QApplication(sys.argv)
fm1=face_make()
sys.exit(app.exec_())
4 视频演示：
dlib人脸合成演示