使用InsightFace识别人脸
Face-Recognition-with-InsightFace
MTCNN
InsightFace
pip install keras
pip install tensorflow==2.2
1
2
datasets\train里的目录名中的空格’ ‘改成’_’
使用datasets\train里的图片生成train.lst
修改src\data\dir2lst.py
import sys
import os
sys.path.append('../common')
import face_image
input_dir = sys.argv[1]
dataset = face_image.get_dataset_common(input_dir, 2)
output_filename = os.path.join(input_dir, 'train.lst')
with open(output_filename, "a") as text_file:
for item in dataset:
oline = "%d\t%s\t%d\n" % (1, item.image_path, int(item.classname))
text_file.write(oline)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
python dir2lst.py F:\insightface-master0\datasets\train
1
修改src\data\face2rec2.py
#try:
#import multiprocessing #Windows下
#except ImportError:
s = mx.recordio.pack(header, b'')#加上b
1
2
3
4
5
6
python face2rec2.py F:\insightface-master0\datasets\train
1
2
使用generate_image_pairs.py
并修改
if len(same_list) > 10: #and len(same_list) < 13
for j in range(0, 10, 2): #len(same_list)
1
2
python generate_image_pairs.py --data-dir F:\insightface-master0\datasets\train --outputtxt F:\insightface-master0\datasets\train\train.txt --num-samepairs 100
1
修改lfw2pack.py
python lfw2pack2.py --data-dir F:\insightface-master0\datasets\train --output F:\insightface-master0\datasets\train\train.bin --num-samepairs 100
1
生成的文件放到recognition\datasets\train里
新建文件property
内容为类别数,112,112
recognition\ArcFace
复制sample_config.py为config.py
修改config.py
dataset.emore.dataset = 'emore'
dataset.emore.dataset_path = '../datasets/train'
dataset.emore.num_classes = #类别数#
dataset.emore.image_shape = (112, 112, 3)
dataset.emore.val_targets = ['train']
default.end_epoch = 100
default.per_batch_size = 32 #128# 显卡垃圾
1
2
3
4
5
6
7
8
9
修改verification.py
#val = float(true_accept) / float(n_same)
#far = float(false_accept) / float(n_diff)
#改成
if n_same == 0:
val = 1
else:
val = float(true_accept) / float(n_same)
if n_diff == 0:
far = 0
else:
far = float(false_accept) / float(n_diff)
1
2
3
4
5
6
7
8
9
10
11
12
修改train.py
#_rescale = 1.0 / args.ctx_num
_rescale = 0.03125
1
2
然后训练并验证
set CUDA_VISIBLE_DEVICES='0,'
python -u train.py --network r100 --loss arcface --dataset emore
1
2
3
修改src\train_softmax.py
#print(his.history['accuracy'])
history['acc'] += his.history['accuracy']
history['val_acc'] += his.history['val_accuracy']
1
2
3
4
参考RetinaFace\test.py修改src\recognizer_image.py
from retinaface import RetinaFace
...
detector = RetinaFace('../RetinaFace/model/R50', 0, gpuid, 'net3')
...
faces, landmarks = detector.detect(img, thresh, scales=scales, do_flip=flip)
...
1
2
3
4
5
6
python faces_embedding.py --dataset F:\insightface-master0\datasets\train
python train_softmax.py
python recognizer_image.py --image-in ../datasets/test/005.jpg
1
2
3
4
其它
下载lfw-deepfunneled
使用src\align\align_lfw.py 生成对齐后的人脸
或参考编译RetinaFace及使用
_paths = fimage.image_path.split('\\')#Windows下'/'改成'\\'
1
python align_lfw.py --input-dir F:\lfw-deepfunneled --output-dir F:\lfw-align
1
最终
用insightface检测新浪微博下载的图片并识别
import cv2
import sys
import numpy as np
import datetime
import os
import glob
from skimage import transform as trans
sys.path.append('../deploy')
sys.path.append('../src/common')
sys.path.append('../RetinaFace')
from retinaface import RetinaFace
from keras.models import load_model
#from mtcnn.mtcnn import MTCNN
from imutils import paths
import face_preprocess
import numpy as np
import face_model
import argparse
import pickle
import time
import cv2
import os
ap = argparse.ArgumentParser()
ap.add_argument("--mymodel", default="outputs/my_model.h5",
help="Path to recognizer model")
ap.add_argument("--le", default="outputs/le.pickle",
help="Path to label encoder")
ap.add_argument("--embeddings", default="outputs/embeddings.pickle",
help='Path to embeddings')
ap.add_argument('--image-size', default='112,112', help='')
ap.add_argument('--model', default='../models/model-y1-test2/model,0', help='path to load model.')
ap.add_argument('--ga-model', default='', help='path to load model.')
ap.add_argument('--gpu', default=0, type=int, help='gpu id')
ap.add_argument('--det', default=0, type=int, help='mtcnn option, 1 means using R+O, 0 means detect from begining')
ap.add_argument('--flip', default=0, type=int, help='whether do lr flip aug')
ap.add_argument('--threshold', default=1.24, type=float, help='ver dist threshold')
args = ap.parse_args()
# Load embeddings and labels
data = pickle.loads(open(args.embeddings, "rb").read())
le = pickle.loads(open(args.le, "rb").read())
embeddings = np.array(data['embeddings'])
labels = le.fit_transform(data['names'])
# Initialize faces embedding model
embedding_model = face_model.FaceModel(args)
# Load the classifier model
model = load_model(args.mymodel)
gpuid = -1 ###-1禁止使用GPU
detector = RetinaFace('../RetinaFace/model/R50', 0, gpuid, 'net3')
# Setup some useful arguments
cosine_threshold = 0.8
proba_threshold = 0.85
comparing_num = 5
thresh = 0.8
#from face_preprocess
def preprocess(img, bbox=None, landmark=None, **kwargs):
M = None
image_size = [112,112]
if landmark is not None:
assert len(image_size)==2
src = np.array([
[30.2946, 51.6963],
[65.5318, 51.5014],
[48.0252, 71.7366],
[33.5493, 92.3655],
[62.7299, 92.2041] ], dtype=np.float32 )
if image_size[1]==112:
src[:,0] += 8.0
dst = landmark.astype(np.float32)
tform = trans.SimilarityTransform()
tform.estimate(dst, src)
M = tform.params[0:2,:]
#M = cv2.estimateRigidTransform( dst.reshape(1,5,2), src.reshape(1,5,2), False)
if M is None:
if bbox is None: #use center crop
det = np.zeros(4, dtype=np.int32)
det[0] = int(img.shape[1]*0.0625)
det[1] = int(img.shape[0]*0.0625)
det[2] = img.shape[1] - det[0]
det[3] = img.shape[0] - det[1]
else:
det = bbox
margin = kwargs.get('margin', 44)
bb = np.zeros(4, dtype=np.int32)
bb[0] = np.maximum(det[0]-margin/2, 0)
bb[1] = np.maximum(det[1]-margin/2, 0)
bb[2] = np.minimum(det[2]+margin/2, img.shape[1])
bb[3] = np.minimum(det[3]+margin/2, img.shape[0])
ret = img[bb[1]:bb[3],bb[0]:bb[2],:]
if len(image_size)>0:
ret = cv2.resize(ret, (image_size[1], image_size[0]))
return ret
else: #do align using landmark
assert len(image_size)==2
#print(src.shape, dst.shape)
#print(src)
#print(dst)
#print(M)
warped = cv2.warpAffine(img,M,(image_size[1],image_size[0]), borderValue = 0.0)
#tform3 = trans.ProjectiveTransform()
#tform3.estimate(src, dst)
#warped = trans.warp(img, tform3, output_shape=_shape)
return warped
# Define distance function
def findCosineDistance(vector1, vector2):
"""
Calculate cosine distance between two vector
"""
vec1 = vector1.flatten()
vec2 = vector2.flatten()
a = np.dot(vec1.T, vec2)
b = np.dot(vec1.T, vec1)
c = np.dot(vec2.T, vec2)
return 1 - (a/(np.sqrt(b)*np.sqrt(c)))
def CosineSimilarity(test_vec, source_vecs):
"""
Verify the similarity of one vector to group vectors of one class
"""
cos_dist = 0
for source_vec in source_vecs:
cos_dist += findCosineDistance(test_vec, source_vec)
return cos_dist/len(source_vecs)
# 读取中文路径
def cv_imread(filePath):
cv_img=cv2.imdecode(np.fromfile(filePath,dtype=np.uint8),-1)
if cv_img is None:
return cv_img
if len(cv_img.shape) == 2:
cv_img=cv2.cvtColor(cv_img,cv2.COLOR_GRAY2BGR)
return cv_img
def detect(count, jpgfile, spath):
print(jpgfile)
img = cv_imread(jpgfile)
if img is None:
return
index = jpgfile.rfind('.')
if index > 0:
suf = jpgfile[index:]
else:
suf='.jpg'
print(img.shape)
scales = [1024, 1980]
im_shape = img.shape
target_size = scales[0]
max_size = scales[1]
im_size_min = np.min(im_shape[0:2])
im_size_max = np.max(im_shape[0:2])
#im_scale = 1.0
#if im_size_min>target_size or im_size_max>max_size:
im_scale = float(target_size) / float(im_size_min)
# prevent bigger axis from being more than max_size:
if np.round(im_scale * im_size_max) > max_size:
im_scale = float(max_size) / float(im_size_max)
print('im_scale', im_scale)
scales = [im_scale]
flip = False
faces, landmarks = detector.detect(img, thresh, scales=scales, do_flip=flip)
print(count, faces.shape, landmarks.shape)
#print(type(faces))
#print(type(landmarks))
if faces is not None:
print('find', faces.shape[0], 'faces')
for i in range(faces.shape[0]):
#print('score', faces[i][4])
box = faces[i].astype(np.int)
if (box[3]-box[1]) > 100 and (box[2]-box[0]) > 100:
crop = img[box[1]:box[3], box[0]:box[2]]
nimg = preprocess(img, bbox=box, landmark = landmarks[i])#, image_size='112,112'
nimg = cv2.cvtColor(nimg, cv2.COLOR_BGR2RGB)
nimg = np.transpose(nimg, (2,0,1))
embedding = embedding_model.get_feature(nimg).reshape(1,-1)
text = "Unknown"
# Predict class
preds = model.predict(embedding)
preds = preds.flatten()
# Get the highest accuracy embedded vector
j = np.argmax(preds)
proba = preds[j]
# Compare this vector to source class vectors to verify it is actual belong to this class
match_class_idx = (labels == j)
match_class_idx = np.where(match_class_idx)[0]
selected_idx = np.random.choice(match_class_idx, comparing_num)
compare_embeddings = embeddings[selected_idx]
# Calculate cosine similarity
cos_similarity = CosineSimilarity(embedding, compare_embeddings)
if cos_similarity < cosine_threshold and proba > proba_threshold:
name = le.classes_[j]
text = "{}".format(name)
print("Recognized: {} <{:.2f}>".format(name, proba*100))
if text == 'yz':
target_file = os.path.join(spath, str(count)+'__'+str(i)+suf)
cv2.imwrite(target_file, crop)
#oline = '%d\t%s\t%d\n' % (1,target_file, 1)#one class
#text_file.write(oline)
img = None
count=0
ppath="G:\\down\\yz"
spath="G:\\down\\detect_yz"
dirlist=os.listdir(ppath)
for dirs in dirlist:
Olddir=os.path.join(ppath, dirs)
if os.path.isdir(Olddir):
output_filename = os.path.join(spath, 'lst')
npath = os.path.join(spath, dirs[0:4])
isExists = os.path.exists(npath)
if not isExists:
os.makedirs(npath)
filelist1=os.listdir(Olddir)
#with open(output_filename, "a") as text_file:
for files1 in filelist1:
oldfile=os.path.join(Olddir, files1)
detect(count, oldfile, npath)
count+=1