def demo_using_images():
img = Image.open('images/0_fp_1_aligned.jpg')
bounding_boxes, landmarks = detect_faces(img)
img = cv.cvtColor(np.array(img), cv.COLOR_RGB2BGR)
show_bboxes(img, bounding_boxes, landmarks)
img = Image.open('images/0_fp_0_aligned.jpg')
bounding_boxes, landmarks = detect_faces(img)
img = cv.cvtColor(np.array(img), cv.COLOR_RGB2BGR)
show_bboxes(img, bounding_boxes, landmarks)
img = Image.open('images/0_fn_0_aligned.jpg')
bounding_boxes, landmarks = detect_faces(img)
img = cv.cvtColor(np.array(img), cv.COLOR_RGB2BGR)
show_bboxes(img, bounding_boxes, landmarks)
img = Image.open('images/5_fn_0_aligned.jpg')
bounding_boxes, landmarks = detect_faces(img)
img = cv.cvtColor(np.array(img), cv.COLOR_RGB2BGR)
show_bboxes(img, bounding_boxes, landmarks)
img = Image.open('images/6_fn_1_aligned.jpg')
bounding_boxes, landmarks = detect_faces(img)
img = cv.cvtColor(np.array(img), cv.COLOR_RGB2BGR)
show_bboxes(img, bounding_boxes, landmarks)
def image_loader(image_path, out_size, device, transform=None):
global dst_img
raw = cv.imread(image_path)
img = Image.open(image_path).convert('RGB')
_, facial5points = detect_faces(img)
if len(facial5points) != 1:
raise Exception("No face or multi faces...")
facial5points = np.reshape(facial5points[0], (2, 5))
# Default set
crop_size = 224
inner_padding_factor = 0.1
outer_padding = 0
output_size = 224
# get the reference 5 landmarks position in the crop settings
reference_5pts = get_reference_facial_points(
(output_size, output_size), inner_padding_factor,
(outer_padding, outer_padding), True)
dst_img = warp_and_crop_face(raw,
facial5points,
reference_pts=reference_5pts,
crop_size=(crop_size, crop_size))
dst_img = cv.resize(dst_img, (out_size, out_size))[:, :, ::-1]
dst_img_ = Image.fromarray(dst_img)
im = transform(dst_img_).float()
return im.unsqueeze(0).to(device)
def camera(plfd_backbone, video_path, is_video):
transform = transforms.Compose([transforms.ToTensor()])
if is_video:
cap = cv2.VideoCapture(video_path)
else:
cap = cv2.VideoCapture(0)
while True:
ret, img = cap.read()
if not ret: break
height, width = img.shape[:2]
bounding_boxes, landmarks = detect_faces(img)
for box in bounding_boxes:
score = box[4]
x1, y1, x2, y2 = (box[:4] + 0.5).astype(np.int32)
w = x2 - x1 + 1
h = y2 - y1 + 1
size = int(max([w, h]) * 1.1)
cx = x1 + w // 2
cy = y1 + h // 2
x1 = cx - size // 2
x2 = x1 + size
y1 = cy - size // 2
y2 = y1 + size
dx = max(0, -x1)
dy = max(0, -y1)
x1 = max(0, x1)
y1 = max(0, y1)
edx = max(0, x2 - width)
edy = max(0, y2 - height)
x2 = min(width, x2)
y2 = min(height, y2)
cropped = img[y1:y2, x1:x2]
if (dx > 0 or dy > 0 or edx > 0 or edy > 0):
cropped = cv2.copyMakeBorder(cropped, dy, edy, dx, edx,
cv2.BORDER_CONSTANT, 0)
cropped = cv2.resize(cropped, (112, 112))
input = cv2.resize(cropped, (112, 112))
input = cv2.cvtColor(input, cv2.COLOR_BGR2RGB)
input = transform(input).unsqueeze(0).cuda(0)
with torch.no_grad():
landmarks, _ = plfd_backbone(input)
pre_landmark = landmarks[0]
pre_landmark = pre_landmark.cpu().detach().numpy().reshape(
-1, 2) * [size, size]
for (x, y) in pre_landmark.astype(np.int32):
cv2.circle(img, (x1 + x, y1 + y), 0, (0, 0, 255), 2)
cv2.rectangle(img, tuple(box[:2].astype(np.int)),
tuple(box[2:4].astype(np.int)), (0, 0, 255), 1)
cv2.imshow('0', img)
if cv2.waitKey(10) == 27:
break
def auto_anno(dir_path):
for f in os.listdir(dir_path):
p = os.path.join(dir_path, f)
print p
if ".xml" in p:
continue
image = ip.image_to_3channels(p, True)
if image is None:
continue
start = time.clock()
bounding_boxes, landmarks = det.detect_faces(image)
end = time.clock()
print "take %sms" % (1000 * end -
1000 * start), "num faces:", len(bounding_boxes)
if len(bounding_boxes) > 0:
bounding_boxes = [[bbox[0], bbox[1], bbox[2], bbox[3]]
for bbox in bounding_boxes]
bounding_boxes = np.round(bounding_boxes)
bounding_boxes = np.asarray(bounding_boxes, np.int)
ap.save_anno_to_xml(p,
image.size[0],
image.size[1],
3,
annos=bounding_boxes)
def detection(self, imgpath):
img = cv2.imread(imgpath)
bounding_boxes, _ = detect_faces(img)
ul = np.array(bounding_boxes[0, 0:2])
br = np.array(bounding_boxes[0, 2:4])
w = br[0] - ul[0]
h = br[1] - ul[1]
uls = (ul - np.array([w * 0.1, h * 0.1])).astype(np.int)
brs = (br + np.array([w * 0.1, h * 0.1])).astype(np.int)
w, h = brs - uls
if h > w:
diff = h - w
uls[0] = uls[0] - math.floor(diff / 2)
brs[0] = brs[0] + math.ceil(diff / 2)
else:
diff = w - h
uls[1] = uls[1] - math.floor(diff / 2)
brs[1] = brs[1] + math.floor(diff / 2)
self.uls = uls
self.brs = brs
self.box = [*uls, *brs]
self.factor = (self.brs[0] - self.uls[0]) / 112
img112 = cv2.resize(deepcopy(img[uls[1]:brs[1], uls[0]:brs[0], :]), (112, 112), interpolation=cv2.INTER_CUBIC)
return img112
def main(args):
checkpoint = torch.load(args.model_path, map_location=device)
plfd_backbone = PFLDInference().to(device)
plfd_backbone.load_state_dict(checkpoint['plfd_backbone'])
plfd_backbone.eval()
plfd_backbone = plfd_backbone.to(device)
transform = transforms.Compose([transforms.ToTensor()])
cap = cv2.VideoCapture(0)
while True:
ret, img = cap.read()
if not ret: break
height, width = img.shape[:2]
bounding_boxes, landmarks = detect_faces(img)
for box in bounding_boxes:
score = box[4]
x1, y1, x2, y2 = (box[:4] + 0.5).astype(np.int32)
w = x2 - x1 + 1
h = y2 - y1 + 1
size = int(max([w, h]) * 1.1)
cx = x1 + w // 2
cy = y1 + h // 2
x1 = cx - size // 2
x2 = x1 + size
y1 = cy - size // 2
y2 = y1 + size
dx = max(0, -x1)
dy = max(0, -y1)
x1 = max(0, x1)
y1 = max(0, y1)
edx = max(0, x2 - width)
edy = max(0, y2 - height)
x2 = min(width, x2)
y2 = min(height, y2)
cropped = img[y1:y2, x1:x2]
if dx > 0 or dy > 0 or edx > 0 or edy > 0:
cropped = cv2.copyMakeBorder(cropped, dy, edy, dx, edx,
cv2.BORDER_CONSTANT, 0)
cropped = cv2.resize(cropped, (112, 112))
input = cv2.resize(cropped, (112, 112))
input = cv2.cvtColor(input, cv2.COLOR_BGR2RGB)
input = transform(input).unsqueeze(0).to(device)
_, landmarks = plfd_backbone(input)
pre_landmark = landmarks[0]
pre_landmark = pre_landmark.cpu().detach().numpy().reshape(
-1, 2) * [size, size]
for (x, y) in pre_landmark.astype(np.int32):
cv2.circle(img, (x1 + x, y1 + y), 1, (0, 0, 255))
cv2.imshow('0', img)
if cv2.waitKey(10) == 27:
break
def get_faces_mtcnn(self, img_bgr):
"""
基于MTCNN检测人脸
"""
from mtcnn.detector import detect_faces
bbox_mtcnn, landmarks_mtcnn = detect_faces(img_bgr, self.mtcnn_pro, min_face_size=50)
bbox_list, lms_list, size_list = [], [], []
for bbox, lms in zip(bbox_mtcnn, landmarks_mtcnn):
box_prob = bbox[4] # 置信度
if box_prob < 0.9: # 小于0.9直接pass
continue
box = bbox[0:4].astype(np.int32)
bbox_list.append(box)
lms_tmp = lms.astype(np.int32)
lms_points = []
for x, y in zip(lms_tmp[0:5], lms_tmp[5:10]):
lms_points.append([x, y])
lms_list.append(lms_points)
# draw_box(img_bgr, box, is_new=False)
# draw_points(img_bgr, lms_points, is_new=False)
# from root_dir import IMGS_DIR
# img_path = os.path.join(IMGS_DIR, "mtcnn_res.jpg")
# cv2.imwrite(img_path, img_bgr)
return bbox_list, lms_list
def main(args):
checkpoint = torch.load(args.model_path, map_location=device)
pfld_backbone = PFLDInference().to(device)
pfld_backbone.load_state_dict(checkpoint['plfd_backbone'])
pfld_backbone.eval()
pfld_backbone = pfld_backbone.to(device)
transform = transforms.Compose([transforms.ToTensor()])
im = Image.open(args.image_path)
img = np.array(im)
height, width = img.shape[:2]
draw = ImageDraw.Draw(im)
bounding_boxes, landmarks = detect_faces(img)
print(bounding_boxes)
for box in bounding_boxes:
score = box[4]
x1, y1, x2, y2 = (box[:4] + 0.5).astype(np.int32)
w = x2 - x1 + 1
h = y2 - y1 + 1
size = int(max([w, h]) * 1.1)
cx = x1 + w // 2
cy = y1 + h // 2
x1 = cx - size // 2
x2 = x1 + size
y1 = cy - size // 2
y2 = y1 + size
dx = max(0, -x1)
dy = max(0, -y1)
x1 = max(0, x1)
y1 = max(0, y1)
edx = max(0, x2 - width)
edy = max(0, y2 - height)
x2 = min(width, x2)
y2 = min(height, y2)
cropped = img[y1:y2, x1:x2]
if (dx > 0 or dy > 0 or edx > 0 or edy > 0):
cropped = cv2.copyMakeBorder(cropped, dy, edy, dx, edx,
cv2.BORDER_CONSTANT, 0)
cropped = cv2.resize(cropped, (112, 112))
input = cv2.resize(cropped, (112, 112))
input = cv2.cvtColor(input, cv2.COLOR_BGR2RGB)
input = transform(input).unsqueeze(0).to(device)
landmarks = pfld_backbone(input)
pre_landmark = landmarks[0]
pre_landmark = pre_landmark.cpu().detach().numpy().reshape(
-1, 2) * [size, size]
print(pre_landmark)
for (x, y) in pre_landmark.astype(np.int32):
# cv2.circle(img, (x1 + x, y1 + y), 1, (0, 0, 255))
draw.ellipse((x1 + x - 1, y1 + y - 1, x1 + x + 1, y1 + y + 1),
fill=(0, 0, 255))
im.show()
def get_central_face_attributes(full_path):
img = Image.open(full_path).convert('RGB')
bounding_boxes, landmarks = detect_faces(img)
if len(landmarks) > 0:
i = select_central_face(img.size, bounding_boxes)
return True, [bounding_boxes[i]], [landmarks[i]]
return False, None, None
def get_faces_v2(self, face_dict):
"""
基于MTCNN检测人脸
:param face_dict: 数据字典
:return: 人脸检测框
"""
img_op = face_dict['img']
bounding_boxes, landmarks = detect_faces(img_op)
return bounding_boxes
def get_all_face_attributes(full_path):
try:
img = Image.open(full_path).convert('RGB')
bounding_boxes, landmarks = detect_faces(img)
if len(landmarks) > 0:
return True, bounding_boxes, landmarks
except KeyboardInterrupt:
raise
except:
pass
return False, None, None
def extract_keypoints(img_path, model_path):
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
checkpoint = torch.load(model_path, map_location=device)
pfld_backbone = PFLDInference().to(device)
pfld_backbone.load_state_dict(checkpoint['pfld_backbone'])
pfld_backbone.eval()
pfld_backbone = pfld_backbone.to(device)
transform = torchvision.transforms.Compose(
[torchvision.transforms.ToTensor()])
img = cv2.imread(img_path, 1)
height, width = img.shape[:2]
bounding_boxes, landmarks = detect_faces(img)
for box in bounding_boxes:
x1, y1, x2, y2 = (box[:4] + 0.5).astype(np.int32)
w = x2 - x1 + 1 # 宽度
h = y2 - y1 + 1 # 高度
cx = x1 + w // 2 # 中心宽度
cy = y1 + h // 2 # 中心高度
size = int(max([w, h]) * 1.1)
x1 = cx - size // 2
x2 = x1 + size
y1 = cy - size // 2
y2 = y1 + size
x1 = max(0, x1)
y1 = max(0, y1)
x2 = min(width, x2)
y2 = min(height, y2)
edx1 = max(0, -x1)
edy1 = max(0, -y1)
edx2 = max(0, x2 - width)
edy2 = max(0, y2 - height)
cropped = img[y1:y2, x1:x2]
if edx1 > 0 or edy1 > 0 or edx2 > 0 or edy2 > 0:
cropped = cv2.copyMakeBorder(cropped, edy1, edy2, edx1, edx2,
cv2.BORDER_CONSTANT, 0)
input = cv2.resize(cropped, (112, 112))
cv2.imwrite(img_path.replace(".png", "_1.png"), input)
input = transform(input).unsqueeze(0).to(device)
_, landmarks = pfld_backbone(input)
pre_landmark = landmarks[0]
key_points = pre_landmark.cpu().detach().numpy().reshape(
-1, 2) * [size, size] - [edx1, edy1]
keypoints = []
for (x, y) in key_points:
cv2.circle(img, (x1 + int(x), y1 + int(y)), 2, (0, 255, 0), -1)
keypoints.append([x1 + int(x), y1 + int(y)])
cv2.imshow('face_landmark_68', img)
cv2.waitKey(1000)
return keypoints
def get_face_all_attributes(full_path):
try:
img = Image.open(full_path).convert('RGB')
bounding_boxes, landmarks = detect_faces(img)
if len(landmarks) > 0:
i = select_central_face(img.size, bounding_boxes)
return True, [bounding_boxes[i]], [landmarks[i]]
except KeyboardInterrupt:
raise
except:
pass
return False, None, None
def predict_landmarks(self, img_op):
height, width = img_op.shape[:2]
bounding_boxes, landmarks = detect_faces(img_op) # 检测人脸
res_list = []
for box in bounding_boxes:
score = box[4]
x1, y1, x2, y2 = (box[:4] + 0.5).astype(np.int32)
w = x2 - x1 + 1
h = y2 - y1 + 1
size = int(max([w, h]) * 1.1)
cx = x1 + w // 2
cy = y1 + h // 2
x1 = cx - size // 2
x2 = x1 + size
y1 = cy - size // 2
y2 = y1 + size
dx = max(0, -x1)
dy = max(0, -y1)
x1 = max(0, x1)
y1 = max(0, y1)
edx = max(0, x2 - width)
edy = max(0, y2 - height)
x2 = min(width, x2)
y2 = min(height, y2)
cropped = img_op[y1:y2, x1:x2]
if dx > 0 or dy > 0 or edx > 0 or edy > 0:
cropped = cv2.copyMakeBorder(cropped, dy, edy, dx, edx,
cv2.BORDER_CONSTANT, 0)
cropped = cv2.resize(cropped, (112, 112))
input = cv2.resize(cropped, (112, 112))
input = cv2.cvtColor(input, cv2.COLOR_BGR2RGB)
input = self.transform(input).unsqueeze(0).to(self.device)
_, landmarks = self.plfd_backbone(input)
pre_landmark = landmarks[0]
pre_landmark = pre_landmark.cpu().detach().numpy().reshape(
-1, 2) * [size, size]
res_list.append([pre_landmark, x1, y1]) # 添加
return res_list
def main(args):
checkpoint = torch.load(args.model_path)
plfd_backbone = PFLDInference().cuda()
plfd_backbone.load_state_dict(checkpoint['plfd_backbone'])
plfd_backbone.eval()
plfd_backbone = plfd_backbone.cuda()
transform = transforms.Compose([transforms.ToTensor()])
cap = cv2.VideoCapture(0)
while True:
ret, img = cap.read()
if not ret: break
bounding_boxes, landmarks = detect_faces(img)
maxval_index = 0
if (len(bounding_boxes) > 0):
maxval_index = np.argmax(bounding_boxes[:, 4])
if (len(bounding_boxes) > maxval_index - 1
and len(bounding_boxes) > 0):
face = img[int(bounding_boxes[maxval_index][1]
):int(bounding_boxes[maxval_index][3]),
int(bounding_boxes[maxval_index][0]
):int(bounding_boxes[maxval_index][2])]
if face.size == 0:
continue
face_resized = cv2.resize(face,
dsize=(112, 112),
interpolation=cv2.INTER_LINEAR)
face_resized = cv2.cvtColor(face_resized, cv2.COLOR_BGR2RGB)
face_resized = transform(face_resized).unsqueeze(0).cuda()
_, landmarks = plfd_backbone(face_resized)
pre_landmark = landmarks.cpu().detach().numpy()[0].reshape(
-1, 2) * [
int(bounding_boxes[maxval_index][2]) -
int(bounding_boxes[maxval_index][0]),
int(bounding_boxes[maxval_index][3]) -
int(bounding_boxes[maxval_index][1])
]
for (x, y) in pre_landmark.astype(np.int32):
cv2.circle(face, (x, y), 1, (255, 0, 0), -1)
cv2.imshow("xxxx", face)
if cv2.waitKey(10) == 27:
break
def setUp(self):
img_fn = 'images/0_img.jpg'
self.image = cv.imread(img_fn, True)
img = Image.open(img_fn).convert('RGB')
_, facial5points = detect_faces(img)
facial5points = np.reshape(facial5points[0], (2, 5))
self.facial5points = facial5points
default_square = True
inner_padding_factor = 0
outer_padding = (0, 0)
output_size = (112, 112)
self.reference_5pts = get_reference_facial_points(
output_size, inner_padding_factor, outer_padding, default_square)
self.output_size = output_size
def detect_test(path):
for f in os.listdir(path):
p = os.path.join(path, f)
image = ip.image_to_3channels(p, True)
if image is None:
continue
start = time.clock()
bounding_boxes, landmarks = det.detect_faces(image)
end = time.clock()
print "take %sms" % (1000 * end -
1000 * start), "num faces:", len(bounding_boxes)
if len(bounding_boxes) > 0:
draw = ImageDraw.Draw(image)
for bbox in bounding_boxes:
bbox = (bbox[0], bbox[1], bbox[2], bbox[3])
draw.rectangle(bbox, outline="red")
plt.imshow(image)
plt.show()
def demo_using_webcam():
cam = VideoCapture(0) # set the port of the camera as before
cam.set(cv2.CAP_PROP_FPS, 15)
cam.set(cv2.CAP_PROP_FRAME_WIDTH, 640)
cam.set(cv2.CAP_PROP_FRAME_HEIGHT, 480)
cv2.namedWindow("image", cv2.WND_PROP_FULLSCREEN)
cv2.setWindowProperty("image", cv2.WND_PROP_FULLSCREEN,
cv2.WINDOW_FULLSCREEN)
while True:
ret_val, img = cam.read(
) # return a True boolean and and the image if all go right
if ret_val:
bounding_boxes, landmarks = detect_faces(Image.fromarray(img))
show_bboxes(img, bounding_boxes, landmarks, wait=False)
else:
break
cam.release() # Closes video file or capturing device.
def get_face_attributes(full_path):
try:
img = Image.open(full_path).convert('RGB')
bounding_boxes, landmarks = detect_faces(img)
width, height = img.size
if len(bounding_boxes) == 1:
x1, y1, x2, y2 = bounding_boxes[0][0], bounding_boxes[0][
1], bounding_boxes[0][2], bounding_boxes[0][3]
if x1 < 0 or x1 >= width or x2 < 0 or x2 >= width or y1 < 0 or y1 >= height or y2 < 0 or y2 >= height or x1 >= x2 or y1 >= y2:
return False, None, None
landmarks = [int(round(x)) for x in landmarks[0]]
is_valid = (x2 - x1) > width / 10 and (y2 - y1) > height / 10
return is_valid, (int(round(x1)), int(round(y1)), int(round(x2)),
int(round(y2))), landmarks
except KeyboardInterrupt:
raise
except:
pass
return False, None, None
def get_your_faceImg_feature(name, model, features_dict):
faceImg_path = './video/your_face_img'
num_img_useful = 0
for idx, img_name in enumerate(os.listdir(faceImg_path)):
# detect face in your img
img = Image.open(faceImg_path + '/' + img_name).convert('RGB')
try:
bounding_boxes, landmarks = detect_faces(img,
pnet=pnet,
rnet=rnet,
onet=onet,
device=device)
except:
continue
if len(bounding_boxes) == 0:
continue
# crop the face from the img (112, 112)
face_img, box = crop_face(np.array(img), bounding_boxes, landmarks)
feature = get_face_feature(model, face_img)
features_dict[name + '_' + str(idx)] = feature
num_img_useful += 1
print('Successfully collected your face img for %ds ' % num_img_useful)
return features_dict
def face_detect():
start = time.time()
ensure_folder(STATIC_DIR)
ensure_folder(UPLOAD_DIR)
file = request.files['file']
fn = secure_filename(file.filename)
full_path = os.path.join(UPLOAD_DIR, fn)
file.save(full_path)
# resize(full_path)
print('full_path: ' + full_path)
img = Image.open(full_path).convert('RGB')
bboxes, landmarks = detect_faces(img)
num_faces = len(bboxes)
if num_faces > 0:
img = cv.imread(full_path)
draw_bboxes(img, bboxes, landmarks)
cv.imwrite(full_path, img)
elapsed = time.time() - start
return num_faces, float(elapsed), str(fn), bboxes, landmarks
def show_img(self):
global temp_t
success, self.img = self.camera.read()
if success:
self.Image_num += 1
if self.Image_num % 10 == 9:
frame_rate = 10 / (time.clock() - self.timelb)
self.FmRateLCD.display(frame_rate)
self.timelb = time.clock()
if self.case == 0:
showImg = cv2.cvtColor(self.img, cv2.COLOR_BGR2RGB)
showImg = qimage2ndarray.array2qimage(showImg)
self.Camera_2.setPixmap(QPixmap(showImg)) # 展示图片
self.Camera_2.show()
if self.case == 1:
bounding_boxes, landmarks = detect_faces(self.img)
self.img = show_bboxes(self.img, bounding_boxes, landmarks)
showImg = cv2.cvtColor(self.img, cv2.COLOR_BGR2RGB)
showImg = qimage2ndarray.array2qimage(showImg)
self.Camera_2.setPixmap(QPixmap(showImg)) # 展示图片
self.Camera_2.show()
if self.case == 2:
img_copy = self.img.copy()
frag_gray = False
self.time_ing = time.time()
# point=[100,0,540,480]
if self.frag_cap:
bounding_boxes, landmarks = detect_faces(self.img)
print('正在定位······')
if len(bounding_boxes) == 1:
self.point.clear()
for b in bounding_boxes:
b = [int(round(value)) for value in b]
for i in b:
self.point.append(i)
self.frag_cap = False
# print(point)
# cv2.rectangle(draw, (b[0], b[1]), (b[2], b[3]), (0, 255, 0), 2)
# 裁剪坐标为[y0:y1, x0:x1]
if not self.frag_cap:
if self.point[0] < 540:
self.img = self.img[self.point[1] - 10:479,
self.point[0] - 100:self.point[2] +
100]
else:
self.img = self.img[self.point[1] - 10:479,
self.point[0] - 100:639]
else:
self.img = self.img[1:479, 1:640]
if int(self.time_ing - self.time_first) % 60 == 0:
self.frag_cap = True
else:
self.frag_cap = False
bounding_boxes, landmarks = detect_faces(self.img)
#通过MTCNN人脸框判断，当检测不到人脸时判断低头or瞌睡
if len(bounding_boxes) == 0:
self.nod_fps += 1
if self.nod_fps >= 3:
self.Head_state.setText('点头')
self.nod_count += 1
if len(bounding_boxes) > 0:
self.nod_fps = 0
#通过头部姿态欧拉角角度变化判断是否摇头
if len(bounding_boxes) > 0:
Head_Y_X_Z = get_head_pose(landmarks)
print('pitch:{}, yaw:{}, roll:{}'.format(
Head_Y_X_Z[1], Head_Y_X_Z[2], Head_Y_X_Z[3]))
if (Head_Y_X_Z[2] < -0.75):
self.shake_fps_l += 1
if (Head_Y_X_Z[2] >= -0.75):
self.shake_fps_l = 0
if self.shake_fps_l >= 5:
self.shake_count += 1
self.Head_state.setText('摇头')
if Head_Y_X_Z[3] >= 0.30:
self.shake_fps_r += 1
if self.shake_fps_r >= 5:
self.shake_count += 1
self.Head_state.setText('摇头')
if Head_Y_X_Z[3] < 0.30:
self.shake_fps_r = 0
# print(Head_Y_X_Z[1])
# print(Head_Y_X_Z[2])
# print(Head_Y_X_Z[3])
if time.time() - self.nod_start > 3:
self.Head_state.setText('')
if time.time() - self.shake_start > 3:
self.Head_state.setText('')
# 计算低头频率 每10s计算一次
if time.time() - self.nod_start > 10:
times = time.time() - self.nod_start
self.nod_freq = self.nod_count / times
self.nod_start = time.time()
self.Nod_LCD.display(self.nod_freq)
# 计算摇头频率
if time.time() - self.shake_start > 10:
times = time.time() - self.shake_start
self.shake_freq = self.shake_count / times
self.shake_start = time.time()
self.shake_LCD.display(self.shake_freq)
if len(bounding_boxes) > 0:
Emotions = get_emotion(
get_face_expression(self.img, bounding_boxes))
self.Emotion.setText(Emotions[1])
self.Emotion_pred.display(float(Emotions[0]))
# print(Emotions)
canvas = cv2.imread('img_resource/label_pred.jpg',
flags=cv2.IMREAD_UNCHANGED)
for (i,
(emotion,
prob)) in enumerate(zip(self.EMOTIONS, Emotions[2])):
# text = "{}: {:.2f}%".format(emotion, prob * 100)
text = "{:.2f}%".format(prob * 100)
# 绘制表情类和对应概率的条形图
w = int(prob * 180)
# print(text)
# canvas = 255 * np.ones((250, 300, 3), dtype="uint8")
cv2.rectangle(canvas, (0, (i * 44) + 25),
(w, (i * 43) + 40), (100, 200, 130), -1)
cv2.putText(canvas, text, (170, (i * 43) + 40),
cv2.FONT_HERSHEY_DUPLEX, 0.6, (0, 0, 0), 1)
show = cv2.cvtColor(canvas, cv2.COLOR_BGR2RGB)
showImage = QtGui.QImage(show.data, show.shape[1],
show.shape[0],
QtGui.QImage.Format_RGB888)
# cv2.imshow('test', showImage)
# showImg=QPixmap(showImage)
self.label_pred_img.setPixmap(
QtGui.QPixmap.fromImage(showImage))
# # print('test')
# print('Head_Y_X_Z')
# print(Head_Y_X_Z)
x = cv2.resize(self.img, (300, 300)).astype(np.float32)
flag_B = True # 是否闭眼的flag
flag_Y = False
num_rec = 0 # 检测到的眼睛的数量
# 分界线
x -= self.img_mean
x = x.astype(np.float32)
x = x[:, :, ::-1].copy()
x = torch.from_numpy(x).permute(2, 0, 1)
xx = Variable(x.unsqueeze(0))
# if torch.cuda.is_available():
# xx = xx.cuda()
xx = xx.cuda()
y = self.net(xx)
softmax = nn.Softmax(dim=-1)
detect = Detect(config.class_num, 0, 200, 0.01, 0.45)
priors = utils.default_prior_box()
loc, conf = y
loc = torch.cat([o.view(o.size(0), -1) for o in loc], 1)
conf = torch.cat([o.view(o.size(0), -1) for o in conf], 1)
detections = detect(
loc.view(loc.size(0), -1, 4),
softmax(conf.view(conf.size(0), -1, config.class_num)),
torch.cat([o.view(-1, 4) for o in priors], 0)).data
labels = VOC_CLASSES
# 将检测结果放置于图片上
scale = torch.Tensor(self.img.shape[1::-1]).repeat(2)
self.img = show_bboxes(self.img, bounding_boxes, landmarks)
for i in range(detections.size(1)):
j = 0
while detections[0, i, j, 0] >= 0.4:
score = detections[0, i, j, 0]
label_name = labels[i - 1]
if label_name == 'calling' and score > 0.8:
self.Danger_state.setText('打电话')
self.danger_count += 1
frag_gray = True
if label_name == 'smoke' and score > 0.8:
self.Danger_state.setText('吸烟')
self.danger_count += 1
frag_gray = True
if label_name != 'smoke' and label_name != 'calling':
self.danger_t += 1
if self.danger_t >= 20:
self.Danger_state.setText('')
self.danger_t = 0
if label_name == 'open_eye':
self.open_t += 1
if self.open_t >= 20:
self.Eyes_state.setText('')
self.open_t = 0
if label_name == 'closed_mouth':
self.Mouth_state.setText(' ')
if label_name == 'closed_eye':
flag_B = False
frag_gray = True
if label_name == 'open_mouth':
flag_Y = True
display_txt = '%s:%.2f' % (label_name, score)
pt = (detections[0, i, j, 1:] * scale).cpu().numpy()
self.coords = (
pt[0], pt[1]), pt[2] - pt[0] + 1, pt[3] - pt[1] + 1
color = self.colors_tableau[i]
cv2.rectangle(self.img, (pt[0], pt[1]), (pt[2], pt[3]),
color, 2)
cv2.putText(self.img, display_txt,
(int(pt[0]), int(pt[1]) + 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.4,
(255, 255, 255), 1, 8)
j += 1
num_rec += 1
# cv2.imshow('test', self.img)
if num_rec > 0:
if flag_B:
# print(' 1:eye-open')
self.list_B = np.append(self.list_B, 1) # 睁眼为‘1’
self.list_blink = np.append(self.list_blink, 1)
else:
# print(' 0:eye-closed')
self.list_B = np.append(self.list_B, 0) # 闭眼为‘0’
self.list_blink = np.append(self.list_blink, 0)
self.list_blink = np.delete(self.list_blink, 0)
self.list_B = np.delete(self.list_B, 0)
if flag_Y:
self.list_Y = np.append(self.list_Y, 1)
else:
self.list_Y = np.append(self.list_Y, 0)
self.list_Y = np.delete(self.list_Y, 0)
else:
self.Msg.clear()
self.Msg.setPlainText('Nothing detected.')
# print(list)
# 实时计算PERCLOS
self.perclos = 1 - np.average(self.list_blink)
# print('perclos={:f}'.format(perclos))
self.PERCLOS.display(self.perclos)
if self.list_B[8] == 1 and self.list_B[9] == 0:
# 如果上一帧为’1‘，此帧为’0‘则判定为眨眼
self.Eyes_state.setText('眨眼')
self.blink_count += 1
frag_gray = True
str = datetime.datetime.now().strftime("%H:%M:%S")
self.State_record.append(str + '：眨眼')
# img_copy=cv2.cvtColor(img_copy,cv2.COLOR_RGB2GRAY)
blink_T = time.time() - self.blink_start
if blink_T > 30:
# 每30秒计算一次眨眼频率
blink_freq = self.blink_count / blink_T
self.blink_start = time.time()
self.blink_count = 0
print('blink_freq={:f}'.format(blink_freq))
self.Blink_freq.display(blink_freq * 2)
# 检测打哈欠
# if Yawn(list_Y,list_Y1):
if (self.list_Y[len(self.list_Y) -
len(self.list_Y1):] == self.list_Y1).all():
# print('----------------------打哈欠----------------------')
self.Mouth_state.setText('打哈欠')
self.yawn_count += 1
frag_gray = True
str = datetime.datetime.now().strftime("%H:%M:%S")
self.State_record.append(str + '：打哈欠')
self.list_Y = np.zeros(50)
# 计算打哈欠频率
yawn_T = time.time() - self.yawn_start
if yawn_T > 60:
yawn_freq = self.yawn_count / yawn_T
self.yawn_start = time.time()
self.yawn_count = 0
print('yawn_freq={:f}'.format(yawn_freq))
self.Yawn_freq.display(yawn_freq)
# 计算危险行为频率
DangerAct_T = time.time() - self.danger_start
if DangerAct_T > 60:
danger_freq = self.danger_count / DangerAct_T
self.danger_start = time.time()
self.danger_count = 0
print('danger_freq={:f}'.format(danger_freq))
self.Danger_LCD.display(danger_freq)
if (self.perclos > 0.4):
# print('疲劳')
self.State.setText('疲劳')
elif (self.blink_freq > 0.3):
# print('疲劳')
self.State.setText('疲劳')
self.blink_freq = 0 # 如果因为眨眼频率判断疲劳，则初始化眨眼频率
elif (self.yawn_freq > 5.0 / 60):
# print("疲劳")
self.State.setText('疲劳')
self.yawn_freq = 0 # 初始化，同上
else:
self.State.setText('清醒')
if not frag_gray:
showImg = cv2.cvtColor(img_copy, cv2.COLOR_BGR2RGB)
else:
if self.isRecordImg:
str = datetime.datetime.now().strftime(
"%Y_%m_%d_%H_%M_%S")
temp = 'ImgRecord/' + str + '.jpg'
cv2.imwrite(temp, img_copy)
showImg = cv2.cvtColor(img_copy, cv2.COLOR_RGB2GRAY)
showImg = qimage2ndarray.array2qimage(showImg)
self.Camera_2.setPixmap(QPixmap(showImg)) # 展示图片
self.Camera_2.show()
if self.case == 3:
img_copy = self.img.copy()
frag_gray = False
self.time_ing = time.time()
# point=[100,0,540,480]
if self.frag_cap:
bounding_boxes, landmarks = detect_faces(self.img)
print('正在定位······')
if len(bounding_boxes) == 1:
self.point.clear()
for b in bounding_boxes:
b = [int(round(value)) for value in b]
for i in b:
self.point.append(i)
self.frag_cap = False
# print(point)
# cv2.rectangle(draw, (b[0], b[1]), (b[2], b[3]), (0, 255, 0), 2)
# 裁剪坐标为[y0:y1, x0:x1]
if not self.frag_cap:
if self.point[0] < 540:
self.img = self.img[self.point[1] - 10:479,
self.point[0] - 100:self.point[2] +
100]
else:
self.img = self.img[self.point[1] - 10:479,
self.point[0] - 100:639]
else:
self.img = self.img[1:479, 1:640]
if int(self.time_ing - self.time_first) % 60 == 0:
self.frag_cap = True
else:
self.frag_cap = False
bounding_boxes, landmarks = detect_faces(self.img)
# 通过MTCNN人脸框判断，当检测不到人脸时判断低头or瞌睡
if len(bounding_boxes) == 0:
self.nod_fps += 1
if self.nod_fps >= 3:
self.Head_state.setText('点头')
self.nod_count += 1
if len(bounding_boxes) > 0:
self.nod_fps = 0
# 通过头部姿态欧拉角角度变化判断是否摇头
if len(bounding_boxes) > 0:
Head_Y_X_Z = get_head_pose(landmarks)
print('pitch:{}, yaw:{}, roll:{}'.format(
Head_Y_X_Z[1], Head_Y_X_Z[2], Head_Y_X_Z[3]))
if (Head_Y_X_Z[2] < -0.75):
self.shake_fps_l += 1
if (Head_Y_X_Z[2] >= -0.75):
self.shake_fps_l = 0
if self.shake_fps_l >= 5:
self.shake_count += 1
self.Head_state.setText('摇头')
if Head_Y_X_Z[3] >= 0.30:
self.shake_fps_r += 1
if self.shake_fps_r >= 5:
self.shake_count += 1
self.Head_state.setText('摇头')
if Head_Y_X_Z[3] < 0.30:
self.shake_fps_r = 0
# print(Head_Y_X_Z[1])
# print(Head_Y_X_Z[2])
# print(Head_Y_X_Z[3])
if time.time() - self.nod_start > 3:
self.Head_state.setText('')
if time.time() - self.shake_start > 3:
self.Head_state.setText('')
# 计算低头频率 每10s计算一次
if time.time() - self.nod_start > 10:
times = time.time() - self.nod_start
self.nod_freq = self.nod_count / times
self.nod_start = time.time()
self.Nod_LCD.display(self.nod_freq)
# 计算摇头频率
if time.time() - self.shake_start > 10:
times = time.time() - self.shake_start
self.shake_freq = self.shake_count / times
self.shake_start = time.time()
self.shake_LCD.display(self.shake_freq)
if len(bounding_boxes) > 0:
Emotions = get_emotion(
get_face_expression(self.img, bounding_boxes))
self.Emotion.setText(Emotions[1])
self.Emotion_pred.display(float(Emotions[0]))
# print(Emotions)
canvas = cv2.imread('img_resource/label_pred.jpg',
flags=cv2.IMREAD_UNCHANGED)
for (i,
(emotion,
prob)) in enumerate(zip(self.EMOTIONS, Emotions[2])):
# text = "{}: {:.2f}%".format(emotion, prob * 100)
text = "{:.2f}%".format(prob * 100)
# 绘制表情类和对应概率的条形图
w = int(prob * 180)
# print(text)
# canvas = 255 * np.ones((250, 300, 3), dtype="uint8")
cv2.rectangle(canvas, (0, (i * 44) + 25),
(w, (i * 43) + 40), (100, 200, 130), -1)
cv2.putText(canvas, text, (170, (i * 43) + 40),
cv2.FONT_HERSHEY_DUPLEX, 0.6, (0, 0, 0), 1)
show = cv2.cvtColor(canvas, cv2.COLOR_BGR2RGB)
showImage = QtGui.QImage(show.data, show.shape[1],
show.shape[0],
QtGui.QImage.Format_RGB888)
# cv2.imshow('test', showImage)
# showImg=QPixmap(showImage)
self.label_pred_img.setPixmap(
QtGui.QPixmap.fromImage(showImage))
# # print('test')
# print('Head_Y_X_Z')
# print(Head_Y_X_Z)
x = cv2.resize(self.img, (300, 300)).astype(np.float32)
flag_B = True # 是否闭眼的flag
flag_Y = False
num_rec = 0 # 检测到的眼睛的数量
# 分界线
x -= self.img_mean
x = x.astype(np.float32)
x = x[:, :, ::-1].copy()
x = torch.from_numpy(x).permute(2, 0, 1)
xx = Variable(x.unsqueeze(0))
# if torch.cuda.is_available():
# xx = xx.cuda()
xx = xx.cuda()
y = self.net(xx)
softmax = nn.Softmax(dim=-1)
detect = Detect(config.class_num, 0, 200, 0.01, 0.45)
priors = utils.default_prior_box()
loc, conf = y
loc = torch.cat([o.view(o.size(0), -1) for o in loc], 1)
conf = torch.cat([o.view(o.size(0), -1) for o in conf], 1)
detections = detect(
loc.view(loc.size(0), -1, 4),
softmax(conf.view(conf.size(0), -1, config.class_num)),
torch.cat([o.view(-1, 4) for o in priors], 0)).data
labels = VOC_CLASSES
# 将检测结果放置于图片上
scale = torch.Tensor(self.img.shape[1::-1]).repeat(2)
self.img = show_bboxes(self.img, bounding_boxes, landmarks)
for i in range(detections.size(1)):
j = 0
while detections[0, i, j, 0] >= 0.4:
score = detections[0, i, j, 0]
label_name = labels[i - 1]
if label_name == 'calling' and score > 0.8:
self.Danger_state.setText('打电话')
self.danger_count += 1
frag_gray = True
if label_name == 'smoke' and score > 0.8:
self.Danger_state.setText('吸烟')
self.danger_count += 1
frag_gray = True
if label_name != 'smoke' and label_name != 'calling':
self.danger_t += 1
if self.danger_t >= 20:
self.Danger_state.setText('')
self.danger_t = 0
if label_name == 'open_eye':
self.open_t += 1
if self.open_t >= 20:
self.Eyes_state.setText('')
self.open_t = 0
if label_name == 'closed_mouth':
self.Mouth_state.setText(' ')
if label_name == 'closed_eye':
flag_B = False
frag_gray = True
if label_name == 'open_mouth':
flag_Y = True
display_txt = '%s:%.2f' % (label_name, score)
pt = (detections[0, i, j, 1:] * scale).cpu().numpy()
self.coords = (
pt[0], pt[1]), pt[2] - pt[0] + 1, pt[3] - pt[1] + 1
color = self.colors_tableau[i]
cv2.rectangle(self.img, (pt[0], pt[1]), (pt[2], pt[3]),
color, 2)
cv2.putText(self.img, display_txt,
(int(pt[0]), int(pt[1]) + 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.4,
(255, 255, 255), 1, 8)
j += 1
num_rec += 1
# cv2.imshow('test', self.img)
if num_rec > 0:
if flag_B:
# print(' 1:eye-open')
self.list_B = np.append(self.list_B, 1) # 睁眼为‘1’
self.list_blink = np.append(self.list_blink, 1)
else:
# print(' 0:eye-closed')
self.list_B = np.append(self.list_B, 0) # 闭眼为‘0’
self.list_blink = np.append(self.list_blink, 0)
self.list_blink = np.delete(self.list_blink, 0)
self.list_B = np.delete(self.list_B, 0)
if flag_Y:
self.list_Y = np.append(self.list_Y, 1)
else:
self.list_Y = np.append(self.list_Y, 0)
self.list_Y = np.delete(self.list_Y, 0)
else:
self.Msg.clear()
self.Msg.setPlainText('Nothing detected.')
# print(list)
# 实时计算PERCLOS
self.perclos = 1 - np.average(self.list_blink)
# print('perclos={:f}'.format(perclos))
self.PERCLOS.display(self.perclos)
if self.list_B[8] == 1 and self.list_B[9] == 0:
# 如果上一帧为’1‘，此帧为’0‘则判定为眨眼
self.Eyes_state.setText('眨眼')
self.blink_count += 1
frag_gray = True
str = datetime.datetime.now().strftime("%H:%M:%S")
self.State_record.append(str + '：眨眼')
# img_copy=cv2.cvtColor(img_copy,cv2.COLOR_RGB2GRAY)
blink_T = time.time() - self.blink_start
if blink_T > 30:
# 每30秒计算一次眨眼频率
blink_freq = self.blink_count / blink_T
self.blink_start = time.time()
self.blink_count = 0
print('blink_freq={:f}'.format(blink_freq))
self.Blink_freq.display(blink_freq * 2)
# 检测打哈欠
# if Yawn(list_Y,list_Y1):
if (self.list_Y[len(self.list_Y) -
len(self.list_Y1):] == self.list_Y1).all():
# print('----------------------打哈欠----------------------')
self.Mouth_state.setText('打哈欠')
self.yawn_count += 1
frag_gray = True
str = datetime.datetime.now().strftime("%H:%M:%S")
self.State_record.append(str + '：打哈欠')
self.list_Y = np.zeros(50)
# 计算打哈欠频率
yawn_T = time.time() - self.yawn_start
if yawn_T > 60:
yawn_freq = self.yawn_count / yawn_T
self.yawn_start = time.time()
self.yawn_count = 0
print('yawn_freq={:f}'.format(yawn_freq))
self.Yawn_freq.display(yawn_freq)
# 计算危险行为频率
DangerAct_T = time.time() - self.danger_start
if DangerAct_T > 60:
danger_freq = self.danger_count / DangerAct_T
self.danger_start = time.time()
self.danger_count = 0
print('danger_freq={:f}'.format(danger_freq))
self.Danger_LCD.display(danger_freq)
if (self.perclos > 0.4):
# print('疲劳')
self.State.setText('疲劳')
elif (self.blink_freq > 0.3):
# print('疲劳')
self.State.setText('疲劳')
self.blink_freq = 0 # 如果因为眨眼频率判断疲劳，则初始化眨眼频率
elif (self.yawn_freq > 5.0 / 60):
# print("疲劳")
self.State.setText('疲劳')
self.yawn_freq = 0 # 初始化，同上
else:
self.State.setText('清醒')
if not frag_gray:
showImg = cv2.cvtColor(img_copy, cv2.COLOR_BGR2RGB)
else:
if self.isRecordImg:
str = datetime.datetime.now().strftime(
"%Y_%m_%d_%H_%M_%S")
temp = 'ImgRecord/' + str + '.jpg'
cv2.imwrite(temp, img_copy)
showImg = cv2.cvtColor(img_copy, cv2.COLOR_RGB2GRAY)
self.State_record.moveCursor(QTextCursor.End)
showImg = qimage2ndarray.array2qimage(showImg)
self.Camera_2.setPixmap(QPixmap(showImg)) # 展示图片
self.Camera_2.show()
cv.putText(img_raw, text, (cx, cy),
cv.FONT_HERSHEY_DUPLEX, 0.5, (255, 255, 255))
# landms
landms = landmarks[i]
cv.circle(img_raw, (landms[0], landms[5]), 1, (0, 0, 255), 4)
cv.circle(img_raw, (landms[1], landms[6]), 1, (0, 255, 255), 4)
cv.circle(img_raw, (landms[2], landms[7]), 1, (255, 0, 255), 4)
cv.circle(img_raw, (landms[3], landms[8]), 1, (0, 255, 0), 4)
cv.circle(img_raw, (landms[4], landms[9]), 1, (255, 0, 0), 4)
# save image
cv.imwrite('images/result.jpg', img_raw)
cv.imshow('image', img_raw)
cv.waitKey(0)
if __name__ == "__main__":
full_path = 'test/Jason Behr_27968.JPG'
img = Image.open(full_path).convert('RGB')
bboxes, landmarks = mtcnn.detect_faces(img)
print(bboxes)
print(landmarks)
show_bboxes(full_path, bboxes, landmarks)
bboxes, landmarks = retinaface.detect_faces(img)
print(bboxes)
print(landmarks)
show_bboxes(full_path, bboxes, landmarks)
def get_all_face_attributes(full_path):
img = Image.open(full_path).convert('RGB')
bounding_boxes, landmarks = detect_faces(img)
return bounding_boxes, landmarks
def face_attributes():
start = time.time()
ensure_folder(STATIC_DIR)
ensure_folder(UPLOAD_DIR)
file = request.files['file']
fn = secure_filename(file.filename)
full_path = os.path.join(UPLOAD_DIR, fn)
file.save(full_path)
# resize(full_path)
print('full_path: ' + full_path)
img = Image.open(full_path).convert('RGB')
bboxes, landmarks = detect_faces(img)
result = None
if len(bboxes) > 0:
i = select_central_face((im_size, im_size), bboxes)
bbox = bboxes[i]
img = cv.imread(full_path)
boxed = draw_bboxes(img, [bbox], [landmarks[i]])
cv.imwrite(full_path, boxed)
img = crop_image(img, bbox)
img = cv.resize(img, (im_size, im_size))
img = transforms.ToPILImage()(img)
img = transformer(img)
img = img.to(device)
inputs = torch.zeros([1, 3, im_size, im_size], dtype=torch.float)
inputs[0] = img
with torch.no_grad():
reg_out, expression_out, gender_out, glasses_out, race_out = model(
inputs)
reg_out = reg_out.cpu().numpy()
age_out = reg_out[0, 0]
pitch_out = reg_out[0, 1]
roll_out = reg_out[0, 2]
yaw_out = reg_out[0, 3]
beauty_out = reg_out[0, 4]
age = int(age_out * 100)
pitch = float('{0:.2f}'.format(pitch_out * 360 - 180))
roll = float('{0:.2f}'.format(roll_out * 360 - 180))
yaw = float('{0:.2f}'.format(yaw_out * 360 - 180))
beauty = float('{0:.2f}'.format(beauty_out * 100))
beauty_prob = float('{0:.4f}'.format(get_prob(beauty)))
_, expression_out = expression_out.topk(1, 1, True, True)
_, gender_out = gender_out.topk(1, 1, True, True)
_, glasses_out = glasses_out.topk(1, 1, True, True)
_, race_out = race_out.topk(1, 1, True, True)
expression_out = expression_out.cpu().numpy()
gender_out = gender_out.cpu().numpy()
glasses_out = glasses_out.cpu().numpy()
race_out = race_out.cpu().numpy()
expression = idx2name(int(expression_out[0, 0]), 'expression')
gender = idx2name(int(gender_out[0, 0]), 'gender')
glasses = idx2name(int(glasses_out[0, 0]), 'glasses')
race = idx2name(int(race_out[0, 0]), 'race')
result = {
'age': age,
'pitch': pitch,
'roll': roll,
'yaw': yaw,
'beauty': beauty,
'beauty_prob': beauty_prob,
'expression': expression,
'gender': gender,
'glasses': glasses,
'race': race
}
elapsed = time.time() - start
return result, float(elapsed), str(fn)
# # 如果连续3次都小于阈值，则表示瞌睡点头一次
# if hCOUNTER >= NOD_AR_CONSEC_FRAMES: # 阈值：3
# hTOTAL += 1
# print('点头')
# # 重置点头帧计数器
# hCOUNTER = 0
##########################################
# # 检测
# time_2=time.time()
# if int(time_2-time_)%20==0:
# frag=True
if frag:
point = [100, 0, 540, 480]
bounding_boxes, landmarks = detect_faces(img)
img = show_bboxes(img, bounding_boxes, landmarks)
if len(bounding_boxes) > 0:
point.clear()
for b in bounding_boxes:
b = [int(round(value)) for value in b]
for i in b:
point.append(i)
frag = False
print(point)
# cv2.rectangle(draw, (b[0], b[1]), (b[2], b[3]), (0, 255, 0), 2)
# 裁剪坐标为[y0:y1, x0:x1]
if frag == False:
img = img[point[1]:480, point[0] - 100:point[2] + 100]
bounding_boxes, landmarks = detect_faces(img)
if len(bounding_boxes) > 0:
# video_src = 0 # camera device id
capture = cv2.VideoCapture(video_src)
if not capture.isOpened():
print('Camera is not opened!')
else:
idx_frame = 0
while True:
ret, frame = capture.read()
idx_frame += 1
if idx_frame % 2 != 0:
continue
idx_frame = 0
cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
start = time.time()
img = Image.fromarray(frame)
try:
bounding_boxes, landmarks = detect_faces(img,
pnet=pnet,
rnet=rnet,
onet=onet,
device=device)
except:
continue
end = time.time()
print(end - start)
frame = show_bboxes(frame, bounding_boxes, landmarks)
cv2.imshow('Video', frame)
cv2.waitKey(1)
capture.release()
from PIL import Image
from mtcnn.detector import detect_faces
import time
camera_id = 0
cap = cv2.VideoCapture(camera_id)
if not cap.isOpened():
sys.exit()
while True:
ret, frame = cap.read()
image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
pilImg = Image.fromarray(numpy.uint8(image))
boxes, landmarks = detect_faces(pilImg)
for box in boxes:
box = [int(i) for i in box]
image = cv2.rectangle(image, (box[0], box[1]), (box[2], box[3]),
(255, 0, 0), 2)
for marks in landmarks:
for i, j in zip(marks[:5], marks[5:]):
image = cv2.circle(image, (int(i), int(j)),
3,
color=(
255,
0,
),
thickness=2)
import cv2 as cv
import numpy as np
from PIL import Image
from align_faces import warp_and_crop_face, get_reference_facial_points
from mtcnn.detector import detect_faces
if __name__ == "__main__":
for i in range(10):
img_fn = 'images/{}_raw.jpg'.format(i)
print('Loading image {}'.format(img_fn))
raw = cv.imread(img_fn, True)
img = Image.open(img_fn).convert('RGB')
_, facial5points = detect_faces(img)
facial5points = np.reshape(facial5points[0], (2, 5))
crop_size = (224, 224)
default_square = True
inner_padding_factor = 0.25
outer_padding = (0, 0)
output_size = (224, 224)
# get the reference 5 landmarks position in the crop settings
reference_5pts = get_reference_facial_points(
output_size, inner_padding_factor, outer_padding, default_square)
# dst_img = warp_and_crop_face(raw, facial5points, reference_5pts, crop_size)
dst_img = warp_and_crop_face(raw, facial5points, reference_pts=reference_5pts, crop_size=crop_size)
cv.imwrite('images/{}_warped.jpg'.format(i), dst_img)
img = cv.resize(raw, (224, 224))
cv.imwrite('images/{}_img.jpg'.format(i), img)
gCategory = ['M', 'W']
if not os.path.exists("results"):
os.mkdir("results")
# (2) Image processing and cropping
for image_path in tqdm.tqdm(sorted(
paths.list_images("./demo"))): # Based on the path of the data
_, name = os.path.split(image_path)
raw = cv.imread(image_path) # BGR
raw = imutils.resize(raw, width=256)
# img = Image.fromarray(raw[:, :, ::-1]) # RGB
t0 = time.time()
boxes, facial5points = detect_faces(raw)
# whether show face keypoints or not
# for p in facial5points:
# for i in range(5):
# cv.circle(raw, (int(p[i]), int(p[i + 5])), 4, (255, 0, 255), -1)
# if len(facial5points) != 1:
# continue
# facial5points = np.reshape(facial5points[0], (2, 5))
# box = boxes[0].astype(int)
if boxes is not None:
for i in range(boxes.shape[0]):
box = boxes[i].astype(np.int)
if facial5points is not None:
facial5points = facial5points[i].astype(np.int)
for l in range(facial5points.shape[0]):