1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
|
import os
from os.path import join
from pathlib import Path
import cv2 as cv
import numpy as np
import imutils
import operator
from app.utils import im_utils, logger_utils
from app.models.bbox import BBox
from app.settings import app_cfg as cfg
from app.settings import types
class DetectorMTCNN:
# https://github.com/ipazc/mtcnn
# pip install mtcnn
dnn_size = (300, 300)
def __init__(self, size=(400,400)):
from mtcnn.mtcnn import MTCNN
self.detector = MTCNN()
def detect(self, im, size=(400,400), conf_thresh=None, pyramids=None, largest=False, zone=None):
'''Detects face using MTCNN and returns (list) of BBox
:param im: (numpy.ndarray) image
:returns list of BBox
'''
bboxes = []
#conf_thresh = self.conf_thresh if conf_thresh is None else conf_thresh
#pyramids = self.pyramids if pyramids is None else pyramids
dnn_size = self.dnn_size if size is None else size
im = im_utils.resize(im, width=dnn_size[0], height=dnn_size[1])
dim = im.shape[:2][::-1]
dets = self.detector.detect_faces(im)
for det in dets:
rect = det['box']
#keypoints = det['keypoints'] # not using here. see 'face_landmarks.py'
bbox = BBox.from_xywh_dim(*rect, dim)
bboxes.append(bbox)
if largest and len(bboxes) > 1:
# only keep largest
bboxes.sort(key=operator.attrgetter('area'), reverse=True)
bboxes = [bboxes[0]]
return bboxes
class DetectorHaar:
im_size = (400, 400)
cascade_name = types.HaarCascade.FRONTAL
def __init__(self, cascade=types.HaarCascade.FRONTAL):
self.log = logger_utils.Logger.getLogger()
def detect(self, im, scale_factor=1.05, overlaps=5):
pass
class DetectorDLIBCNN:
dnn_size = (300, 300)
pyramids = 0
conf_thresh = 0.85
def __init__(self, gpu=0):
import dlib
self.log = logger_utils.Logger.getLogger()
cuda_visible_devices = os.getenv('CUDA_VISIBLE_DEVICES', '')
os.environ['CUDA_VISIBLE_DEVICES'] = str(gpu)
self.log.info('load model: {}'.format(cfg.DIR_MODELS_DLIB_CNN))
self.detector = dlib.cnn_face_detection_model_v1(cfg.DIR_MODELS_DLIB_CNN)
os.environ['CUDA_VISIBLE_DEVICES'] = cuda_visible_devices # reset
def detect(self, im, size=None, conf_thresh=None, pyramids=None, largest=False, zone=None):
bboxes = []
conf_thresh = self.conf_thresh if conf_thresh is None else conf_thresh
pyramids = self.pyramids if pyramids is None else pyramids
dnn_size = self.dnn_size if size is None else size
# resize image
im = im_utils.resize(im, width=dnn_size[0], height=dnn_size[1])
dim = im.shape[:2][::-1]
im = im_utils.bgr2rgb(im) # convert to RGB for dlib
# run detector
mmod_rects = self.detector(im, pyramids)
# sort results
for mmod_rect in mmod_rects:
if mmod_rect.confidence > conf_thresh:
bbox = BBox.from_dlib_dim(mmod_rect.rect, dim)
bboxes.append(bbox)
if zone:
bboxes = [b for b in bboxes if b.cx > zone[0] and b.cx < 1.0 - zone[0] \
and b.cy > zone[1] and b.cy < 1.0 - zone[1]]
if largest and len(bboxes) > 1:
# only keep largest
bboxes.sort(key=operator.attrgetter('area'), reverse=True)
bboxes = [bboxes[0]]
return bboxes
class DetectorDLIBHOG:
size = (320, 240)
pyramids = 0
conf_thresh = 0.85
def __init__(self):
import dlib
self.log = logger_utils.Logger.getLogger()
self.detector = dlib.get_frontal_face_detector()
def detect(self, im, size=None, conf_thresh=None, pyramids=0, largest=False):
conf_thresh = self.conf_thresh if conf_thresh is None else conf_thresh
dnn_size = self.size if size is None else size
pyramids = self.pyramids if pyramids is None else pyramids
im = im_utils.resize(im, width=dnn_size[0], height=dnn_size[1])
dim = im.shape[:2][::-1]
im = im_utils.bgr2rgb(im) # ?
hog_results = self.detector.run(im, pyramids)
bboxes = []
if len(hog_results[0]) > 0:
for rect, score, direction in zip(*hog_results):
if score > conf_thresh:
bbox = BBox.from_dlib_dim(rect, dim)
bboxes.append(bbox)
if largest and len(bboxes) > 1:
# only keep largest
bboxes.sort(key=operator.attrgetter('area'), reverse=True)
bboxes = [bboxes[0]]
return bboxes
class DetectorCVDNN:
dnn_scale = 1.0 # fixed
dnn_mean = (104.0, 177.0, 123.0) # fixed
dnn_crop = False # crop or force resize
size = (300, 300)
conf_thresh = 0.85
def __init__(self):
import dlib
fp_prototxt = join(cfg.DIR_MODELS_CAFFE, 'face_detect', 'opencv_face_detector.prototxt')
fp_model = join(cfg.DIR_MODELS_CAFFE, 'face_detect', 'opencv_face_detector.caffemodel')
self.net = cv.dnn.readNet(fp_prototxt, fp_model)
self.net.setPreferableBackend(cv.dnn.DNN_BACKEND_OPENCV)
self.net.setPreferableTarget(cv.dnn.DNN_TARGET_CPU)
def detect(self, im, size=None, conf_thresh=None, largest=False, pyramids=None):
"""Detects faces and returns (list) of (BBox)"""
conf_thresh = self.conf_thresh if conf_thresh is None else conf_thresh
dnn_size = self.size if size is None else size
im = cv.resize(im, dnn_size)
blob = cv.dnn.blobFromImage(im, self.dnn_scale, dnn_size, self.dnn_mean)
self.net.setInput(blob)
net_outputs = self.net.forward()
bboxes = []
for i in range(0, net_outputs.shape[2]):
conf = net_outputs[0, 0, i, 2]
if conf > conf_thresh:
rect_norm = net_outputs[0, 0, i, 3:7]
bboxes.append(BBox(*rect_norm))
if largest and len(bboxes) > 1:
# only keep largest
bboxes.sort(key=operator.attrgetter('area'), reverse=True)
bboxes = [bboxes[0]]
return bboxes
|
