-
Notifications
You must be signed in to change notification settings - Fork 0
/
03_predict之体感互动.py
279 lines (226 loc) · 10.3 KB
/
03_predict之体感互动.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
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
# -*- codeing = utf-8 -*-
# @Time :2023/3/31 20:18
# @Author :yujunyu
# @Site :
# @File :predict.py
# @software: PyCharm
import cv2
import numpy as np
import os
from matplotlib import pyplot as plt
import time
import mediapipe as mp
import tensorflow as tf
import autopy
import pyautogui
pyautogui.FAILSAFE = False
mp_holistic = mp.solutions.holistic # holistic model
# mp_hands = mp.solutions.hands
mp_drawing = mp.solutions.drawing_utils # drawing utilities
def mediapipe_detection(image, model):
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB) # BGR->RGB
image.flags.writeable = False # image is no longer writeable
results = model.process(image) # make prediction
image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR) # RGB->BGR
return image, results
左: (474, 352)
def draw_styled_landmarks(image, results):
# draw face connections
mp_drawing.draw_landmarks(image, results.face_landmarks, mp_holistic.FACEMESH_CONTOURS,
mp_drawing.DrawingSpec(color=(80, 110, 10), thickness=1, circle_radius=1),
mp_drawing.DrawingSpec(color=(80, 0, 121), thickness=1, circle_radius=1)
)
# draw pose connections
mp_drawing.draw_landmarks(image, results.pose_landmarks, mp_holistic.POSE_CONNECTIONS,
mp_drawing.DrawingSpec(color=(80, 22, 10), thickness=2, circle_radius=4),
mp_drawing.DrawingSpec(color=(80, 44, 121), thickness=2, circle_radius=2)
)
# draw left hand connections
mp_drawing.draw_landmarks(image, results.left_hand_landmarks, mp_holistic.HAND_CONNECTIONS,
mp_drawing.DrawingSpec(color=(121, 22, 76), thickness=2, circle_radius=4),
mp_drawing.DrawingSpec(color=(121, 44, 250), thickness=2, circle_radius=2)
)
# draw right hand connections
mp_drawing.draw_landmarks(image, results.right_hand_landmarks, mp_holistic.HAND_CONNECTIONS,
mp_drawing.DrawingSpec(color=(245, 117, 66), thickness=2, circle_radius=4),
mp_drawing.DrawingSpec(color=(245, 66, 230), thickness=2, circle_radius=2)
)
def extract_keypoints(results):
pose = np.array([[res.x, res.y, res.z, res.visibility] for res in
results.pose_landmarks.landmark]).flatten() if results.pose_landmarks else np.zeros(33 * 4)
face = np.array([[res.x, res.y, res.z] for res in results.face_landmarks.landmark]).flatten() if results.face_landmarks else np.zeros(
468 * 3)
lh = np.array(
[[res.x, res.y, res.z] for res in results.left_hand_landmarks.landmark]).flatten() if results.left_hand_landmarks else np.zeros(
21 * 3)
rh = np.array(
[[res.x, res.y, res.z] for res in results.right_hand_landmarks.landmark]).flatten() if results.right_hand_landmarks else np.zeros(
21 * 3)
#
if results.left_hand_landmarks:
lh_xy_list = []
for index, lm in enumerate(results.left_hand_landmarks.landmark):
lh_xy_list.append([lm.x, lm.y])
else:
lh_xy_list = []
if results.right_hand_landmarks:
rh_xy_list = []
for index, lm in enumerate(results.right_hand_landmarks.landmark):
rh_xy_list.append([lm.x, lm.y])
else:
rh_xy_list = []
return np.concatenate([pose, face, lh, rh]), lh_xy_list, rh_xy_list
def prob_viz(res, actions, input_frame, colors):
output_frame = input_frame.copy()
for num, prob in enumerate(res):
# print(num, prob)
'''
0 0.01494145
1 0.8505417
2 0.13451675
'''
# print(num) # 0 1 2
cv2.rectangle(output_frame, (0, 60 + num * 35), (int(prob * 100), 90 + num * 35), colors[num], -1)
cv2.putText(output_frame, actions[num] + ' {:.2f}'.format(prob), (0, 85 + num * 35), cv2.FONT_HERSHEY_SIMPLEX, 1,
(255, 255, 255), 2, cv2.LINE_AA)
return output_frame
def show_hand_center(image, w, h, lh_xy_list, rh_xy_list):
if len(lh_xy_list) != 0:
# 取hand的9，最靠近中间的坐标
lx, ly = int(lh_xy_list[9][0] * w), int(lh_xy_list[9][1] * h)
print(f'左:{lx, ly}')
cv2.putText(image, 'lh', (lx, ly), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 0), 3)
if len(rh_xy_list) != 0:
rx, ry = int(rh_xy_list[9][0] * w), int(rh_xy_list[9][1] * h)
print(f'右:{rx, ry}')
cv2.putText(image, 'rh', (rx, ry), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 255, 255), 3)
def real_time_predict():
# 1.New detection variables
sequence = []
sentence = []
predictions = []
threshold = 0.4
# Open capture
wCam, hCam = 1920, 1080
frameR = 100
smoothening = 5
plocX, plocY = 0, 0
clocX, clocY = 0, 0
wScr, hScr = autopy.screen.size()
print(wScr, hScr)
cap = cv2.VideoCapture(0)
# cap.set(3, wCam)
# cap.set(4, hCam)
# fps
fps_count = 0
start_time = time.time()
pTime = 0
# Set mediapipe model
with mp_holistic.Holistic(min_detection_confidence=0.5, min_tracking_confidence=0.5) as holistic:
while cap.isOpened():
# Read feed
rer, frame = cap.read()
# filp
# frame = cv2.flip(frame, 1)
# Make detection
image, results = mediapipe_detection(frame, holistic)
# print(results)
# Draw landmarkd
draw_styled_landmarks(image, results)
# 2. Prediction logic
keypoints, lh_xy_list, rh_xy_list = extract_keypoints(results)
# 1)
# sequence.insert(0, keypoints)
# 2) error
# sequence.append(keypoints)
# sequence=sequence[:30]
sequence.append(keypoints)
sequence = sequence[-30:]
# show hand center
h, w, c = image.shape
show_hand_center(image, w, h, lh_xy_list, rh_xy_list)
if len(sequence) == 30:
res = model.predict(np.expand_dims(sequence, axis=0))[0]
# print(res)
print(actions[np.argmax(res)], np.max(res))
predictions.append(np.argmax(res))
### 如果返回某手势，执行指定操作
if actions[np.argmax(res)] == 'normal':
print('这是默认手势，不做任何操作')
elif actions[np.argmax(res)] == 'r_hand':
try:
# 点击鼠标-不释放
autopy.mouse.toggle(None, True)
# pyautogui.mouseDown() # 有延迟，太卡
# 移动鼠标
x_r, y_r = int(rh_xy_list[9][0] * wScr), int(rh_xy_list[9][1] * hScr)
autopy.mouse.move(x_r, y_r)
# pyautogui.moveTo(x_r, y_r, duration=1)
except:
pass
elif actions[np.argmax(res)] == 'a_hand':
try:
x_l, y_l = int(lh_xy_list[9][0] * wScr), int(lh_xy_list[9][1] * hScr)
x_r, y_r = int(rh_xy_list[9][0] * wScr), int(rh_xy_list[9][1] * hScr)
print(x_l - x_r)
if x_l - x_r > 900:
# 释放鼠标
autopy.mouse.toggle(None, False)
# pyautogui.mouseUp()
# 控制滚轮——正：向上，负向下
pyautogui.scroll(200)
# move_len = (x_r - x_l)
autopy.mouse.toggle(None, False)
else:
# 释放鼠标
autopy.mouse.toggle(None, False)
# pyautogui.mouseUp()
# 控制滚轮——正：向上，负向下
pyautogui.scroll(-200)
# move_len = (x_r - x_l)
autopy.mouse.toggle(None, False)
except:
pass
# 3.Viz logic
if np.unique(predictions[-10:])[0] == np.argmax(res):
if res[np.argmax(res)] > threshold:
if len(sentence) > 0:
if actions[np.argmax(res)] != sentence[-1]:
sentence.append(actions[np.argmax(res)])
else:
sentence.append(actions[np.argmax(res)])
if len(sentence) > 5:
sentence = sentence[-5:]
print(sentence[-1])
# 4.Vix probabilities
image = prob_viz(res, actions, image, colors)
cv2.rectangle(image, (0, 0), (640, 40), (245, 117, 16), -1)
cv2.putText(image, ' '.join(sentence), (3, 30), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 255, 255), 2, cv2.LINE_AA)
# # 计算FPS
fps_count += 1
elapsed_time = time.time() - start_time
if elapsed_time > 1: # 每隔1秒钟更新一次FPS
fps = fps_count / elapsed_time
print("\033[33mFPS:{:.2f}\033[0m".format(fps))
fps_count = 0
start_time = time.time()
# FPS
cTime = time.time()
fps = 1 / (cTime - pTime)
pTime = cTime
cv2.putText(image, f'fps:{int(fps)}', (500, 85), cv2.FONT_HERSHEY_SIMPLEX, 1.0, (120, 117, 255), 2, cv2.LINE_AA)
cv2.putText(image, 'exit:q', (500, 120), cv2.FONT_HERSHEY_SIMPLEX, 1.0, (120, 117, 255), 2, cv2.LINE_AA)
# Show to screen
cv2.imshow('Real-Time-Test', image)
# Break
if cv2.waitKey(1) & 0xff == ord('q'):
break
cap.release()
cv2.destroyAllWindows()
if __name__ == '__main__':
import os
os.environ["CUDA_VISIBLE_DEVICES"] = "-1" # 这一行注释掉就是使用gpu，不注释就是使用cpu
actions = np.array(['normal', 'r_hand', 'a_hand'])
model = tf.keras.models.load_model("action-03-tanh.h5")
colors = [(245, 117, 16), (117, 245, 16), (16, 117, 245)]
real_time_predict()