-
Notifications
You must be signed in to change notification settings - Fork 90
/
Copy pathsample_facedetection.py
171 lines (125 loc) · 5.82 KB
/
sample_facedetection.py
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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import copy
import argparse
import cv2 as cv
import numpy as np
import mediapipe as mp
from utils import CvFpsCalc
def get_args():
parser = argparse.ArgumentParser()
parser.add_argument("--device", type=int, default=0)
parser.add_argument("--width", help='cap width', type=int, default=960)
parser.add_argument("--height", help='cap height', type=int, default=540)
parser.add_argument("--model_selection", type=int, default=0)
parser.add_argument("--min_detection_confidence",
help='min_detection_confidence',
type=float,
default=0.7)
parser.add_argument('--use_brect', action='store_true')
args = parser.parse_args()
return args
def main():
# 引数解析 #################################################################
args = get_args()
cap_device = args.device
cap_width = args.width
cap_height = args.height
model_selection = args.model_selection
min_detection_confidence = args.min_detection_confidence
use_brect = args.use_brect
# カメラ準備 ###############################################################
cap = cv.VideoCapture(cap_device)
cap.set(cv.CAP_PROP_FRAME_WIDTH, cap_width)
cap.set(cv.CAP_PROP_FRAME_HEIGHT, cap_height)
# モデルロード #############################################################
mp_face_detection = mp.solutions.face_detection
face_detection = mp_face_detection.FaceDetection(
model_selection=model_selection,
min_detection_confidence=min_detection_confidence,
)
# FPS計測モジュール ########################################################
cvFpsCalc = CvFpsCalc(buffer_len=10)
while True:
display_fps = cvFpsCalc.get()
# カメラキャプチャ #####################################################
ret, image = cap.read()
if not ret:
break
image = cv.flip(image, 1) # ミラー表示
debug_image = copy.deepcopy(image)
# 検出実施 #############################################################
image = cv.cvtColor(image, cv.COLOR_BGR2RGB)
results = face_detection.process(image)
# 描画 ################################################################
if results.detections is not None:
for detection in results.detections:
# 描画
debug_image = draw_detection(debug_image, detection)
cv.putText(debug_image, "FPS:" + str(display_fps), (10, 30),
cv.FONT_HERSHEY_SIMPLEX, 1.0, (0, 255, 0), 2, cv.LINE_AA)
# キー処理(ESC：終了) #################################################
key = cv.waitKey(1)
if key == 27: # ESC
break
# 画面反映 #############################################################
cv.imshow('MediaPipe Face Detection Demo', debug_image)
cap.release()
cv.destroyAllWindows()
def draw_detection(image, detection):
image_width, image_height = image.shape[1], image.shape[0]
# print(detection)
# print(detection.location_data.relative_keypoints[0])
# print(detection.location_data.relative_keypoints[1])
# print(detection.location_data.relative_keypoints[2])
# print(detection.location_data.relative_keypoints[3])
# print(detection.location_data.relative_keypoints[4])
# print(detection.location_data.relative_keypoints[5])
# バウンディングボックス
bbox = detection.location_data.relative_bounding_box
bbox.xmin = int(bbox.xmin * image_width)
bbox.ymin = int(bbox.ymin * image_height)
bbox.width = int(bbox.width * image_width)
bbox.height = int(bbox.height * image_height)
cv.rectangle(image, (int(bbox.xmin), int(bbox.ymin)),
(int(bbox.xmin + bbox.width), int(bbox.ymin + bbox.height)),
(0, 255, 0), 2)
# スコア・ラベルID
cv.putText(
image,
str(detection.label_id[0]) + ":" + str(round(detection.score[0], 3)),
(int(bbox.xmin), int(bbox.ymin) - 20), cv.FONT_HERSHEY_SIMPLEX, 1.0,
(0, 255, 0), 2, cv.LINE_AA)
# キーポイント：右目
keypoint0 = detection.location_data.relative_keypoints[0]
keypoint0.x = int(keypoint0.x * image_width)
keypoint0.y = int(keypoint0.y * image_height)
cv.circle(image, (int(keypoint0.x), int(keypoint0.y)), 5, (0, 255, 0), 2)
# キーポイント：左目
keypoint1 = detection.location_data.relative_keypoints[1]
keypoint1.x = int(keypoint1.x * image_width)
keypoint1.y = int(keypoint1.y * image_height)
cv.circle(image, (int(keypoint1.x), int(keypoint1.y)), 5, (0, 255, 0), 2)
# キーポイント：鼻
keypoint2 = detection.location_data.relative_keypoints[2]
keypoint2.x = int(keypoint2.x * image_width)
keypoint2.y = int(keypoint2.y * image_height)
cv.circle(image, (int(keypoint2.x), int(keypoint2.y)), 5, (0, 255, 0), 2)
# キーポイント：口
keypoint3 = detection.location_data.relative_keypoints[3]
keypoint3.x = int(keypoint3.x * image_width)
keypoint3.y = int(keypoint3.y * image_height)
cv.circle(image, (int(keypoint3.x), int(keypoint3.y)), 5, (0, 255, 0), 2)
# キーポイント：右耳
keypoint4 = detection.location_data.relative_keypoints[4]
keypoint4.x = int(keypoint4.x * image_width)
keypoint4.y = int(keypoint4.y * image_height)
cv.circle(image, (int(keypoint4.x), int(keypoint4.y)), 5, (0, 255, 0), 2)
# キーポイント：左耳
keypoint5 = detection.location_data.relative_keypoints[5]
keypoint5.x = int(keypoint5.x * image_width)
keypoint5.y = int(keypoint5.y * image_height)
cv.circle(image, (int(keypoint5.x), int(keypoint5.y)), 5, (0, 255, 0), 2)
return image
if __name__ == '__main__':
main()