前言

在前面的博客中介紹了，如何使用dlib標定人臉（python dlib學習（一）：人臉檢測），提取68個特征點（python dlib學習（二）：人臉特征點標定）。這次要在這兩個工作的基礎之上，將人臉的信息提取成一個128維的向量空間。在這個向量空間上，同一個人臉的更接近，不同人臉的距離更遠。度量采用歐式距離，歐氏距離計算不算復雜。
二維情況下：

distance=(x1?x2)2+(y1?y2)2??????????????????√
三維情況下：
distance=(x1?x2)2+(y1?y2)2+(z1?z2)2????????????????????????????√
將其擴展到128維的情況下即可。
通常使用的判別閾值是0.6，即如果兩個人臉的向量空間的歐式距離超過了0.6，即認定不是同一個人；如果歐氏距離小于0.6，則認為是同一個人。這個距離也可以由自己定，只要效果能更好。

實驗中使用了兩個模型：

shape_predictor_68_face_landmarks.dat：
http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2

dlib_face_recognition_resnet_model_v1.dat：
http://dlib.net/files/dlib_face_recognition_resnet_model_v1.dat.bz2

文件夾目錄：

兩個模型放在model文件夾中，測試圖片放在faces中，圖片自己隨便下幾張就行。

完整工程下載鏈接：
http://pan.baidu.com/s/1boCDZ7T

程序1

不說廢話了，直接上代碼。

# -*- coding: utf-8 -*- import sys import dlib import cv2 import os import globcurrent_path = os.getcwd() # 獲取當前路徑 # 模型路徑 predictor_path = current_path + "\\model\\shape_predictor_68_face_landmarks.dat" face_rec_model_path = current_path + "\\model\\dlib_face_recognition_resnet_model_v1.dat" #測試圖片路徑 faces_folder_path = current_path + "\\faces\\"# 讀入模型 detector = dlib.get_frontal_face_detector() shape_predictor = dlib.shape_predictor(predictor_path) face_rec_model = dlib.face_recognition_model_v1(face_rec_model_path)for img_path in glob.glob(os.path.join(faces_folder_path, "*.jpg")):print("Processing file: {}".format(img_path))# opencv 讀取圖片，并顯示img = cv2.imread(img_path, cv2.IMREAD_COLOR)# opencv的bgr格式圖片轉換成rgb格式b, g, r = cv2.split(img)img2 = cv2.merge([r, g, b])dets = detector(img, 1) # 人臉標定print("Number of faces detected: {}".format(len(dets)))for index, face in enumerate(dets):print('face {}; left {}; top {}; right {}; bottom {}'.format(index, face.left(), face.top(), face.right(), face.bottom()))shape = shape_predictor(img2, face) # 提取68個特征點for i, pt in enumerate(shape.parts()):#print('Part {}: {}'.format(i, pt))pt_pos = (pt.x, pt.y)cv2.circle(img, pt_pos, 2, (255, 0, 0), 1)#print(type(pt))#print("Part 0: {}, Part 1: {} ...".format(shape.part(0), shape.part(1)))cv2.namedWindow(img_path+str(index), cv2.WINDOW_AUTOSIZE)cv2.imshow(img_path+str(index), img)face_descriptor = face_rec_model.compute_face_descriptor(img2, shape) # 計算人臉的128維的向量print(face_descriptor)k = cv2.waitKey(0) cv2.destroyAllWindows()

程序1結果

部分打印結果：

F:\Python\my_dlib_codes\face_recognition>python my_face_recogniton.py Processing file: F:\Python\my_dlib_codes\face_recognition\faces\jobs.jpg Number of faces detected: 1 face 0; left 184; top 64; right 339; bottom 219 -0.179784 0.15487 0.10509 -0.0973604 -0.19153 0.000418252 -0.0357536 -0.0206766 0.129741 -0.0628359 ....

后面的那一堆數字就是人臉在128維向量空間上的值。

程序2

前面只是測試了一下，把要用的值給求到了。這里我封裝了一下，把比對功能實現了。沒加多少東西，所以不做贅述了。

# -*- coding: utf-8 -*- import sys import dlib import cv2 import os import glob import numpy as npdef comparePersonData(data1, data2):diff = 0# for v1, v2 in data1, data2:# diff += (v1 - v2)**2for i in xrange(len(data1)):diff += (data1[i] - data2[i])**2diff = np.sqrt(diff)print diffif(diff < 0.6):print "It's the same person"else:print "It's not the same person"def savePersonData(face_rec_class, face_descriptor):if face_rec_class.name == None or face_descriptor == None:returnfilePath = face_rec_class.dataPath + face_rec_class.name + '.npy'vectors = np.array([])for i, num in enumerate(face_descriptor):vectors = np.append(vectors, num)# print(num)print('Saving files to :'+filePath)np.save(filePath, vectors)return vectorsdef loadPersonData(face_rec_class, personName):if personName == None:returnfilePath = face_rec_class.dataPath + personName + '.npy'vectors = np.load(filePath)print(vectors)return vectorsclass face_recognition(object):def __init__(self):self.current_path = os.getcwd() # 獲取當前路徑self.predictor_path = self.current_path + "\\model\\shape_predictor_68_face_landmarks.dat"self.face_rec_model_path = self.current_path + "\\model\\dlib_face_recognition_resnet_model_v1.dat"self.faces_folder_path = self.current_path + "\\faces\\"self.dataPath = self.current_path + "\\data\\"self.detector = dlib.get_frontal_face_detector()self.shape_predictor = dlib.shape_predictor(self.predictor_path)self.face_rec_model = dlib.face_recognition_model_v1(self.face_rec_model_path)self.name = Noneself.img_bgr = Noneself.img_rgb = Noneself.detector = dlib.get_frontal_face_detector()self.shape_predictor = dlib.shape_predictor(self.predictor_path)self.face_rec_model = dlib.face_recognition_model_v1(self.face_rec_model_path)def inputPerson(self, name='people', img_path=None):if img_path == None:print('No file!\n')return # img_name += self.faces_folder_path + img_nameself.name = nameself.img_bgr = cv2.imread(self.current_path+img_path)# opencv的bgr格式圖片轉換成rgb格式b, g, r = cv2.split(self.img_bgr)self.img_rgb = cv2.merge([r, g, b])def create128DVectorSpace(self):dets = self.detector(self.img_rgb, 1)print("Number of faces detected: {}".format(len(dets)))for index, face in enumerate(dets):print('face {}; left {}; top {}; right {}; bottom {}'.format(index, face.left(), face.top(), face.right(), face.bottom()))shape = self.shape_predictor(self.img_rgb, face)face_descriptor = self.face_rec_model.compute_face_descriptor(self.img_rgb, shape)# print(face_descriptor)# for i, num in enumerate(face_descriptor):# print(num)# print(type(num))return face_descriptor

程序2結果

測試代碼1：

import face_rec as fc face_rec = fc.face_recognition() # 創建對象 face_rec.inputPerson(name='jobs', img_path='\\faces\\jobs.jpg') # name中寫第一個人名字，img_name為圖片名字，注意要放在faces文件夾中 vector = face_rec.create128DVectorSpace() # 提取128維向量，是dlib.vector類的對象 person_data1 = fc.savePersonData(face_rec, vector ) # 將提取出的數據保存到data文件夾，為便于操作返回numpy數組，內容還是一樣的# 導入第二張圖片，并提取特征向量 face_rec.inputPerson(name='jobs2', img_path='\\faces\\jobs2.jpg') vector = face_rec.create128DVectorSpace() # 提取128維向量，是dlib.vector類的對象 person_data2 = fc.savePersonData(face_rec, vector )# 計算歐式距離，判斷是否是同一個人 fc.comparePersonData(person_data1, person_data2)

如果data文件夾中已經有了模型文件，可以直接導入：

import face_rec as fc face_rec = fc.face_recognition() # 創建對象 person_data1 = fc.loadPersonData(face_rec , 'jobs') # 創建一個類保存相關信息，后面還要跟上人名，程序會在data文件中查找對應npy文件，比如這里就是'jobs.npy' person_data2 = fc.loadPersonData(face_rec , 'jobs2') # 導入第二張圖片 fc.comparePersonData(person_data1, person_data2) # 計算歐式距離，判斷是否是同一個人

程序2結果

Python 2.7.10 |Anaconda 2.3.0 (64-bit)| (default, May 28 2015, 16:44:52) [MSC v.1500 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information. Anaconda is brought to you by Continuum Analytics. Please check out: http://continuum.io/thanks and https://binstar.org >>> import face_rec as fc >>> face_rec = fc.face_recognition() >>> face_rec.inputPerson(name='jobs', img_path='\\faces\\jobs.jpg') >>> vector = face_rec.create128DVectorSpace() Number of faces detected: 1 face 0; left 184; top 64; right 339; bottom 219 >>> person_data1 = fc.savePersonData(face_rec, vector ) Saving files to :F:\Python\my_dlib_codes\face_recognition\data\jobs.npy >>> face_rec.inputPerson(name='jobs2', img_path='\\faces\\jobs2.jpg') >>> vector = face_rec.create128DVectorSpace() Number of faces detected: 1 face 0; left 124; top 39; right 253; bottom 168 >>> person_data2 = fc.savePersonData(face_rec, vector ) Saving files to :F:\Python\my_dlib_codes\face_recognition\data\jobs2.npy >>> fc.comparePersonData(person_data1, person_data2) 0.490491048429 It's the same person

官方例程

#!/usr/bin/python # The contents of this file are in the public domain. See LICENSE_FOR_EXAMPLE_PROGRAMS.txt # # This example shows how to use dlib's face recognition tool. This tool maps # an image of a human face to a 128 dimensional vector space where images of # the same person are near to each other and images from different people are # far apart. Therefore, you can perform face recognition by mapping faces to # the 128D space and then checking if their Euclidean distance is small # enough. # # When using a distance threshold of 0.6, the dlib model obtains an accuracy # of 99.38% on the standard LFW face recognition benchmark, which is # comparable to other state-of-the-art methods for face recognition as of # February 2017. This accuracy means that, when presented with a pair of face # images, the tool will correctly identify if the pair belongs to the same # person or is from different people 99.38% of the time. # # Finally, for an in-depth discussion of how dlib's tool works you should # refer to the C++ example program dnn_face_recognition_ex.cpp and the # attendant documentation referenced therein. # # # # # COMPILING/INSTALLING THE DLIB PYTHON INTERFACE # You can install dlib using the command: # pip install dlib # # Alternatively, if you want to compile dlib yourself then go into the dlib # root folder and run: # python setup.py install # or # python setup.py install --yes USE_AVX_INSTRUCTIONS # if you have a CPU that supports AVX instructions, since this makes some # things run faster. This code will also use CUDA if you have CUDA and cuDNN # installed. # # Compiling dlib should work on any operating system so long as you have # CMake and boost-python installed. On Ubuntu, this can be done easily by # running the command: # sudo apt-get install libboost-python-dev cmake # # Also note that this example requires scikit-image which can be installed # via the command: # pip install scikit-image # Or downloaded from http://scikit-image.org/download.html. import sys import os import dlib import glob from skimage import ioif len(sys.argv) != 4:print("Call this program like this:\n"" ./face_recognition.py shape_predictor_68_face_landmarks.dat dlib_face_recognition_resnet_model_v1.dat ../examples/faces\n""You can download a trained facial shape predictor and recognition model from:\n"" http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2\n"" http://dlib.net/files/dlib_face_recognition_resnet_model_v1.dat.bz2")exit()predictor_path = sys.argv[1] face_rec_model_path = sys.argv[2] faces_folder_path = sys.argv[3]# Load all the models we need: a detector to find the faces, a shape predictor # to find face landmarks so we can precisely localize the face, and finally the # face recognition model. detector = dlib.get_frontal_face_detector() sp = dlib.shape_predictor(predictor_path) facerec = dlib.face_recognition_model_v1(face_rec_model_path)win = dlib.image_window()# Now process all the images for f in glob.glob(os.path.join(faces_folder_path, "*.jpg")):print("Processing file: {}".format(f))img = io.imread(f)win.clear_overlay()win.set_image(img)# 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(img, 1)print("Number of faces detected: {}".format(len(dets)))# Now process each face we found.for k, d in enumerate(dets):print("Detection {}: Left: {} Top: {} Right: {} Bottom: {}".format(k, d.left(), d.top(), d.right(), d.bottom()))# Get the landmarks/parts for the face in box d.shape = sp(img, d)# Draw the face landmarks on the screen so we can see what face is currently being processed.win.clear_overlay()win.add_overlay(d)win.add_overlay(shape)# Compute the 128D vector that describes the face in img identified by# shape. In general, if two face descriptor vectors have a Euclidean# distance between them less than 0.6 then they are from the same# person, otherwise they are from different people. Here we just print# the vector to the screen.face_descriptor = facerec.compute_face_descriptor(img, shape)print(face_descriptor)# It should also be noted that you can also call this function like this:# face_descriptor = facerec.compute_face_descriptor(img, shape, 100)# The version of the call without the 100 gets 99.13% accuracy on LFW# while the version with 100 gets 99.38%. However, the 100 makes the# call 100x slower to execute, so choose whatever version you like. To# explain a little, the 3rd argument tells the code how many times to# jitter/resample the image. When you set it to 100 it executes the# face descriptor extraction 100 times on slightly modified versions of# the face and returns the average result. You could also pick a more# middle value, such as 10, which is only 10x slower but still gets an# LFW accuracy of 99.3%.dlib.hit_enter_to_continue()

吐槽：
dlib的確很方便，不用花多少時間就能自己做到一些目標功能。官方文檔講的很詳細，很容易入門。看這個文檔（dlib python api）差不多就能學會用了。導師已經安排了研究生階段的學習任務了，后面也要忙起來了。dlib的學習雖然是我10月份才開的坑，為了善始善終我也要盡快整理完這些東西。以后要回到”泡館”生活了。
ヽ(･ω･｡)ﾉ

總結

以上是生活随笔為你收集整理的python dlib学习（五）：比对人脸的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。