def main():
dataset = args.dataset
root_dir = args.root_dir
filenamelist = get_list_from_filenames(args.filename_list)
face_detector = FaceDetector(detection_cfg)
count = 0
det_count = 0
faceboxes = {}
if dataset == 'BIWI':
img_ext = '_rgb.png'
else:
img_ext = '.jpg'
for image_name in filenamelist:
image_path = os.path.join(root_dir, image_name + img_ext)
image = cv2.imread(image_path)
bboxes = face_detector.detect_faces(image)
count += 1
if len(bboxes) > 0:
det_count += 1
print("images, %s, detected face %s" % (count, det_count))
x_min, y_min, x_max, y_max = bboxes[0][:4]
xmin = int(x_min)
ymin = int(y_min)
xmax = int(x_max)
ymax = int(y_max)
bbox = [xmin, ymin, xmax, ymax]
faceboxes[image_name] = bbox
# if count >10:
# break
headpose_file = os.path.join('datasets', 'filenamelists',
dataset + '_facebbox_disgard.json')
with open(headpose_file, 'w') as f:
json.dump(faceboxes, f, indent=4)
return
def main():
global face_detector, landmark_dectector, head_pose_estimator, gaze_estimator, mouse_controller
# Load parameters
app_start_time = cv2.getTickCount()
set_logging()
logging.info('start APP')
pyautogui.FAILSAFE = False
cmd_paras = get_comand_line_parameters()
#cmd_paras = get_comand_lineFP16()
# Setup Projects
mouse_controller = createMouseController()
# Setup Classes
logging.info('setUp dectors and estimators started')
# NC2 can only handel one instance of IECore
plugin = IECore()
face_detector = FaceDetector(cmd_paras.fd, cmd_paras.device, plugin)
face_detector.load_model()
landmark_dectector = FacialLandmarksDetector(cmd_paras.lr,
cmd_paras.device, plugin)
landmark_dectector.load_model()
head_pose_estimator = HeadPoseEstimator(cmd_paras.hp, cmd_paras.device,
plugin)
head_pose_estimator.load_model()
gaze_estimator = GazeEstimatior(cmd_paras.ge, cmd_paras.device, plugin)
gaze_estimator.load_model()
logging.info('setUp dectors estimators ends')
logging.info('start inputFeeder read stream')
input_feeder = createInputFeeder(cmd_paras)
input_feeder.load_data()
# RUN the pipline
i = 0
try:
for tmp_image in input_feeder.next_batch():
if tmp_image is not None:
i = i + 1
tmp_image, gaze_position = pipline(tmp_image, i)
cv2.imshow('frame', tmp_image)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
mouse_controller.move(gaze_position[0], gaze_position[1])
else:
break
finally:
input_feeder.close()
logging.info('end inputFeeder end stream')
app_end_time = cv2.getTickCount()
logging.info('App Runtime: {:.4f} seconds pipline runs {:d}'.format(
(app_end_time - app_start_time) / cv2.getTickFrequency(), i))
def __init__(self, video_source=0):
self.COVER_COLOR = (0, 0, 0)
self.detector = FaceDetector()
self.age_estimator = AgeEstimator()
self.capture = cv2.VideoCapture(video_source)
if not self.capture.isOpened():
raise ValueError("Unable to open video source", video_source)
self.width = self.capture.get(cv2.CAP_PROP_FRAME_WIDTH)
self.height = self.capture.get(cv2.CAP_PROP_FRAME_HEIGHT)
def main(args):
print("[INFO] loading facial landmark predictor...")
# Load mtcnn detector from facenet
face_detect = FaceDetector(True, None)
vs, file_stream = getVideoStream(args)
# Read the first frame
success, frame = vs.read()
face_box = None
i = 0
runtime_array = []
face_detection_runtime_array = []
while success:
print("Frame: ", i)
i += 1
frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
# detect faces in the rgb frame
start = time.time()
boxes, _ = face_detect.detectFace(frame_rgb)
stop = time.time()
face_detection_runtime_array.append(stop - start)
frame_draw = frame.copy()
if boxes is not None:
if face_box is None:
face_box = FaceBox(boxes[0], frame, args["shape_predictor"])
else:
face_box.updateFrame(frame)
face_box.updateRect(boxes[0])
start = time.time()
check_liveness = face_box.checkFrame()
stop = time.time()
runtime_array.append(stop - start)
if check_liveness:
print("Real")
break
cv2.imshow("Frame", frame)
key = cv2.waitKey(1) & 0xFF
# if the `q` key was pressed, break from the loop
if key == ord("q"):
break
success, frame = vs.read()
cv2.destroyAllWindows()
if not file_stream:
vs.stop()
sum = 0
for runtime in runtime_array:
sum += runtime
avg = sum / len(runtime_array)
print("Avg blink detection time:", avg)
sum = 0
for runtime in face_detection_runtime_array:
sum += runtime
avg = sum / len(face_detection_runtime_array)
print("Avg face detection time:", avg)
class FaceBlocking:
def __init__(self, video_source=0):
self.COVER_COLOR = (0, 0, 0)
self.detector = FaceDetector()
self.age_estimator = AgeEstimator()
self.capture = cv2.VideoCapture(video_source)
if not self.capture.isOpened():
raise ValueError("Unable to open video source", video_source)
self.width = self.capture.get(cv2.CAP_PROP_FRAME_WIDTH)
self.height = self.capture.get(cv2.CAP_PROP_FRAME_HEIGHT)
def __del__(self):
if self.capture.isOpened():
self.capture.release()
def get_processed_frame(self, _age_restrictions=(), _debug=False):
val, frame = self.capture.read()
faces = self.detector.get_coordinates(frame, _multi_face=True)
faces = faces if faces is not None else []
for i in range(len(faces)):
start_x, start_y, end_x, end_y = faces[i]
roi_color = frame[start_y:end_y, start_x:end_x]
pred_gender, pred_age = self.age_estimator.estimate(roi_color)
estimated_age = np.argmax(pred_age)
if not _age_restrictions[0] <= estimated_age <= _age_restrictions[1]:
self.block_face(frame, start_x, start_y, end_x, end_y)
if _debug:
self.draw_debug(frame, estimated_age, start_x, start_y, end_x, end_y)
return frame
def draw_debug(self, _frame, _age, _start_x, _start_y, _end_x, _end_y):
font = cv2.FONT_HERSHEY_SIMPLEX
color = (0, 255, 0)
stroke = 2
cv2.putText(_frame, "Age: " + str(_age), (_start_x, _start_y), font, 1, color, stroke, cv2.LINE_AA)
cv2.rectangle(_frame, (_start_x, _start_y), (_end_x, _end_y), color, stroke)
def block_face(self, _frame, _start_x, _start_y, _end_x, _end_y):
cv2.rectangle(_frame, (_start_x, _start_y), (_end_x, _end_y), self.COVER_COLOR, -1)
def set_detection_type(self, detection_type):
self.detector.set_detection_type(detection_type)
def set_age_estimation_model(self, estimation_model):
self.age_estimator.switch_model(estimation_model)
def main():
"""Entry point for node."""
# Register ourselves as a node with ROS
rospy.init_node('face_detect')
# Create the face detector
hcd = rospy.get_param('haar_cascade_dir', DEFAULT_HAAR_CASCADE_DIR)
rospy.logdebug('Creating face detector using data in: %s', hcd)
detect_faces = FaceDetector(hcd)
# Create a publisher for detected face results. See [1] for discussion of
# queue_size parameter.
# [1] http://wiki.ros.org/rospy/Overview/Publishers%20and%20Subscribers
faces_pub = rospy.Publisher(rospy.get_name() + '/faces',
Faces,
queue_size=1)
# Create a object encapsulating this node's logic
node = FaceDetectionNode(detect_faces, faces_pub)
# Subscribe to incoming camera images. Note that we set the queue size to
# 1. This means we automatically drop images if we can't detect faces fast
# enough.
rospy.Subscriber('camera/image_raw',
Image,
node.new_input_image,
queue_size=1)
# Run the event loop. Only returns once the node has shutdown.
rospy.spin()
def __init__(self, detection_type, recognition_type, video_source=0):
self.COVER_COLOR = (0, 0, 0)
self.MIN_CONF = 40
self.detector = FaceDetector(detection_type)
self.face_recognizer = FaceRecognizer(recognition_type)
self.embedding_model = load_model(
os.path.join(MODELS_DIR, "facenet_keras.h5"))
self.labels = self.face_recognizer.labels
self.capture = cv2.VideoCapture(video_source)
if not self.capture.isOpened():
raise ValueError("Unable to open video source", video_source)
self.width = self.capture.get(cv2.CAP_PROP_FRAME_WIDTH)
self.height = self.capture.get(cv2.CAP_PROP_FRAME_HEIGHT)
def process_video(
detector: FaceDetector,
filepath: str,
mode: str = 'blur',
):
cap = cv2.VideoCapture(filepath)
_, frame = cap.read()
height, width = frame.shape[:2]
fourcc = cv2.VideoWriter_fourcc(*'XVID')
writer = cv2.VideoWriter('./output/video.avi', fourcc, 20, (width, height), True)
while cap.isOpened():
_, frame = cap.read()
bboxes, landmarks = detector.predict(frame)
if mode == 'blur':
frame = blur(frame, bboxes)
elif mode == 'pixel':
frame = pixelate(frame, bboxes)
else:
frame = hide_eyes(frame, landmarks)
writer.write(frame)
cap.release()
writer.release()
def main():
cfg = anyconfig.load('settings.yaml')
cfg = munch.munchify(cfg)
detector = FaceDetector(cfg.face_detector.model_path, -1)
for ext in exts:
image_list.extend(glob.glob(os.path.join(image_dir, ext)))
for image_path in image_list:
filename = image_path.split('/')[-1] #os should be linux
image = cv2.imread(image_path)
bboxes, facial_lanmarks = detector.detect(image[:, :, ::-1], 1.0)
for i, bbox in enumerate(bboxes):
conf_score = bbox[4]
if conf_score < cfg.face_detector.confident_score_threshold:
continue
xmin, ymin, xmax, ymax = [int(val) for val in bbox[:4]]
# cv2.rectangle(image, (xmin, ymin), (xmax, ymax), (0, 255, 0), 3)
face = image[ymin:ymax, xmin:xmax]
image[ymin:ymax, xmin:xmax] = anonymize_face_pixelate(face)
cv2.imwrite(os.path.join(output_dir, filename), image)
def __init__(self):
self.opts = Options().opts
self.FaceDetector = FaceDetector()
self.interpreter_lm = tf.lite.Interpreter(
model_path=self.opts.landmark_model_path)
self.interpreter_lm.allocate_tensors()
self.input_details_lm = self.interpreter_lm.get_input_details()
self.output_details_lm = self.interpreter_lm.get_output_details()
self.input_shape_lm = self.input_details_lm[0]['shape']
self.output_shape_lm = self.output_details_lm[0]['shape']
self.in_h, self.in_w = self.input_shape_lm[1], self.input_shape_lm[2]
class FaceBlocking:
def __init__(self, detection_type, recognition_type, video_source=0):
self.COVER_COLOR = (0, 0, 0)
self.MIN_CONF = 40
self.detector = FaceDetector(detection_type)
self.face_recognizer = FaceRecognizer(recognition_type)
self.embedding_model = load_model(
os.path.join(MODELS_DIR, "facenet_keras.h5"))
self.labels = self.face_recognizer.labels
self.capture = cv2.VideoCapture(video_source)
if not self.capture.isOpened():
raise ValueError("Unable to open video source", video_source)
self.width = self.capture.get(cv2.CAP_PROP_FRAME_WIDTH)
self.height = self.capture.get(cv2.CAP_PROP_FRAME_HEIGHT)
def __del__(self):
if self.capture.isOpened():
self.capture.release()
def get_processed_frame(self, _block_list=[], _debug=False):
val, frame = self.capture.read()
faces = self.detector.get_coordinates(frame, _multi_face=True)
faces = faces if faces is not None else []
for i in range(len(faces)):
start_x, start_y, end_x, end_y = faces[i]
roi_color = frame[start_y:end_y, start_x:end_x]
who_face, conf = self.face_recognizer.face_classification(
roi_color, self.embedding_model)
if conf >= self.MIN_CONF and who_face in _block_list:
self.block_face(frame, who_face, start_x, start_y, end_x,
end_y)
if _debug:
self.draw_debug(frame, who_face, start_x, start_y, end_x,
end_y)
return frame
def draw_debug(self, _frame, _name, _start_x, _start_y, _end_x, _end_y):
font = cv2.FONT_HERSHEY_SIMPLEX
color = (255, 255, 255)
stroke = 2
cv2.putText(_frame, _name, (_start_x, _start_y), font, 1, color,
stroke, cv2.LINE_AA)
cv2.rectangle(_frame, (_start_x, _start_y), (_end_x, _end_y), color,
stroke)
def block_face(self, _frame, _name, _start_x, _start_y, _end_x, _end_y):
cv2.rectangle(_frame, (_start_x, _start_y), (_end_x, _end_y),
self.COVER_COLOR, -1)
def show_AFLW2000():
face_detector = FaceDetector(detection_cfg)
root_dir = args.root_dir
filenamelist = get_list_from_filenames(args.filename_list)
for image_name in filenamelist:
image_path = os.path.join(root_dir, image_name + ".jpg")
mat_path = os.path.join(root_dir, image_name + '.mat')
# Crop the face loosely
pt2d = utils.get_pt2d_from_mat(mat_path)
x_min = min(pt2d[0, :])
y_min = min(pt2d[1, :])
x_max = max(pt2d[0, :])
y_max = max(pt2d[1, :])
k = 0.20
x_min -= 0.6 * k * abs(x_max - x_min)
y_min -= 2 * k * abs(y_max - y_min)
x_max += 0.6 * k * abs(x_max - x_min)
y_max += 0.6 * k * abs(y_max - y_min)
bboxes = np.array([[x_min, y_min, x_max, y_max, 1.0]])
image = cv2.imread(image_path)
image = face_detector.draw_bboxes(image, bboxes)
cv2.imshow('image', image)
k = cv2.waitKey(500)
return
def __init__(self):
Opts = Options()
opts = Opts.opts
self.recog_th = opts.recog_th
self.face_database = opts.face_database
self.face_model_path = opts.face_recog_model_path
self.feature_extractor = tf.lite.Interpreter(self.face_model_path)
self.feature_extractor.allocate_tensors()
self.input_details = self.feature_extractor.get_input_details()
self.output_details = self.feature_extractor.get_output_details()
self.in_shape = self.input_details[0]['shape']
self.output_shape = self.output_details[0]['shape']
self.FaceDetector = FaceDetector()
self.known_names, self.known_embeddings = self.get_embeddings_dir()
def process_image(
detector: FaceDetector,
filepath: str,
mode: str = 'blur',
):
image = cv2.imread(filepath)
bboxes, landmarks = detector.predict(image)
if mode == 'blur':
image = blur(image, bboxes)
elif mode == 'pixel':
image = pixelate(image, bboxes)
else:
image = hide_eyes(image, landmarks)
os.makedirs('./output', exist_ok=True)
output_path = os.path.join('./output', os.path.split(args.file)[1])
cv2.imwrite(output_path, image)
args = parser.parse_args()
# Set up logging
logging.basicConfig(filename=args.log,
level=logging.INFO,
format="%(asctime)s [%(levelname)s] %(message)s",
datefmt='%Y-%m-%d %I:%M:%S')
logging.info('Initialization...')
try:
feed = InputFeeder(args.input)
t = -time.time() # measure model loading time
faceDetector = FaceDetector(precision=args.precision,
concurrency=args.concurrency,
device=args.device,
extensions=args.ext)
eyeDetector = EyeDetector(precision=args.precision,
concurrency=args.concurrency,
device=args.device,
extensions=args.ext)
headPoseEstimator = HeadPoseEstimator(precision=args.precision,
concurrency=args.concurrency,
device=args.device,
extensions=args.ext)
gazeEstimator = GazeEstimator(precision=args.precision,
concurrency=args.concurrency,
device=args.device,
extensions=args.ext)
mouseController = MouseController(precision='high',
speed=args.speed.lower(),
def infer_on_stream(args):
try:
log.basicConfig(
level=log.INFO,
format="%(asctime)s [%(levelname)s] %(message)s",
handlers=[log.FileHandler("app.log"),
log.StreamHandler()])
mouse_controller = MouseController(precision="low", speed="fast")
start_model_load_time = time.time()
face_detector = FaceDetector(args.model_face_detection)
facial_landmarks_detector = FacialLandmarksDetector(
args.model_facial_landmarks_detection)
head_pose_estimator = HeadPoseEstimator(
args.model_head_pose_estimation)
gaze_estimator = GazeEstimator(args.model_gaze_estimation)
face_detector.load_model()
facial_landmarks_detector.load_model()
head_pose_estimator.load_model()
gaze_estimator.load_model()
total_model_load_time = time.time() - start_model_load_time
log.info("Model load time: {:.1f}ms".format(1000 *
total_model_load_time))
output_directory = os.path.join(args.output_path + '\\' + args.device)
if not os.path.exists(output_directory):
os.makedirs(output_directory)
feed = InputFeeder(args.input_type, args.input_path)
feed.load_data()
out_video = feed.get_out_video(output_directory)
frame_counter = 0
start_inference_time = time.time()
total_prepocess_time = 0
while True:
try:
frame = next(feed.next_batch())
except StopIteration:
break
frame_counter += 1
face_boxes = face_detector.predict(frame)
for face_box in face_boxes:
face_image = get_crop_image(frame, face_box)
eye_boxes, eye_centers = facial_landmarks_detector.predict(
face_image)
left_eye_image, right_eye_image = [
get_crop_image(face_image, eye_box)
for eye_box in eye_boxes
]
head_pose_angles = head_pose_estimator.predict(face_image)
gaze_x, gaze_y = gaze_estimator.predict(
right_eye_image, head_pose_angles, left_eye_image)
draw_gaze_line(frame, face_box, eye_centers, gaze_x, gaze_y)
if args.show_input:
cv2.imshow('im', frame)
if args.move_mouse:
mouse_controller.move(gaze_x, gaze_y)
total_prepocess_time += face_detector.preprocess_time + facial_landmarks_detector.preprocess_time + \
head_pose_estimator.preprocess_time + gaze_estimator.preprocess_time
break
if out_video is not None:
out_video.write(frame)
if args.input_type == "image":
cv2.imwrite(os.path.join(output_directory, 'output_image.jpg'),
frame)
key_pressed = cv2.waitKey(60)
if key_pressed == 27:
break
total_time = time.time() - start_inference_time
total_inference_time = round(total_time, 1)
fps = frame_counter / total_inference_time
log.info("Inference time:{:.1f}ms".format(1000 * total_inference_time))
log.info("Input/output preprocess time:{:.1f}ms".format(
1000 * total_prepocess_time))
log.info("FPS:{}".format(fps))
with open(os.path.join(output_directory, 'stats.txt'), 'w') as f:
f.write(str(total_inference_time) + '\n')
f.write(str(total_prepocess_time) + '\n')
f.write(str(fps) + '\n')
f.write(str(total_model_load_time) + '\n')
feed.close()
cv2.destroyAllWindows()
except Exception as e:
log.exception("Something wrong when running inference:" + str(e))
for subdir in os.listdir(directory):
path = os.path.join(directory, subdir)
# skip any files that might be in the dir
if not os.path.isdir(path):
continue
# load all faces in the subdirectory
faces = get_faces(detector, path)
labels = [subdir.replace("_", " ").title() for _ in range(len(faces))]
print(f'>loaded {len(faces)} examples for person: {subdir.replace("_", " ").title()}')
X_.extend(faces)
y_.extend(labels)
return np.asarray(X_), np.asarray(y_)
if __name__ == '__main__':
detector_base = FaceDetector(detection_type='base')
detector_caffe = FaceDetector(detection_type='caffe')
recognizer_base = BaseRecognizer()
data_base, labels_base = prepare_data_base_recognizer(detector_caffe)
print('Prepared data for base recognizer')
recognizer_base.train_clf(data_base, labels_base)
print('Trained base recognizer')
recognizer_svm = SuppVecMachinesRecognizer()
data_svm, labels_svm = prepare_data_svm_recognizer(detector_caffe, recognizer_svm)
print('Prepared data for svm recognizer')
recognizer_svm.train_clf(data_svm, labels_svm, show_stats=True)
print('Trained svm recognizer')
# for root, dirs, files in os.walk(IMG_DIR):
import cv2
import os
import time
import torch
import argparse
import json
import shutil
import string
from headpose import headpose_estimator
from face_detection import FaceDetector
from face_detection.config import cfg as detection_cfg
face_detector = FaceDetector(detection_cfg)
parser = argparse.ArgumentParser()
parser.add_argument('--root_data', dest='root_data', help='Path to test dataset', default='', type=str)
parser.add_argument('--output_dir', dest='output_dir', help='Path to output', default='', type=str)
args = parser.parse_args()
def get_list_from_filenames(file_path):
# input: relative path to .txt file with file names
# output: list of relative path names
with open(file_path) as f:
lines = f.read().splitlines()
return lines
def main():
print(args)
if mode == 'blur':
frame = blur(frame, bboxes)
elif mode == 'pixel':
frame = pixelate(frame, bboxes)
else:
frame = hide_eyes(frame, landmarks)
writer.write(frame)
cap.release()
writer.release()
if __name__ == '__main__':
parser = ArgumentParser()
parser.add_argument('-f', '--file', type=str, required=True)
parser.add_argument('-m', '--mode', choices=['blur', 'pixel', 'eyes'], default='blur', required=True)
parser.add_argument('-w', '--weights', default='./weights/Resnet50_Final.pth')
args = parser.parse_args()
detector = FaceDetector.from_path(args.weights)
ext = os.path.splitext(args.file)[1]
if ext in IMAGE_FORMATS:
process_image(detector=detector, filepath=args.file, mode=args.mode)
elif ext in VIDEO_FORMATS:
process_video(detector=detector, filepath=args.file, mode=args.mode)
else:
raise Exception(f'Unknown file format: {ext}')
import numpy as np
import cv2
import imageio
import skimage
from face_detection import FaceDetector
import os
from emotion_classifier import EmotionClassifier
os.chdir('..')
face_detector = FaceDetector(minsize=100)
crop_ratio = (0.05, 0.05)
emotion_classifier = EmotionClassifier()
def process_image(image):
if image is None:
return
image = cv2.GaussianBlur(image, (3, 3), 0)
image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
face_bbs, _ = face_detector.detect_face(image_rgb)
for bb in face_bbs:
bb = [int(x) if x > 0 else 0 for x in bb]
face = image[bb[1]:bb[3], bb[0]:bb[2], :]
if face is None:
continue
label_emotion = emotion_classifier.predict(face)
cv2.rectangle(image, (bb[0], bb[1]), (bb[2], bb[3]), (0, 255, 0), 2)
front_scale = (bb[3] - bb[0] + bb[2] - bb[1]) / 130
thickness = int(front_scale / 4) + 1
if thickness < 1:
thickness = 1
def main():
args = build_argparser().parse_args()
input_file = args.input
logger = log.getLogger()
if input_file == "CAM":
input_feeder = InputFeeder("cam")
else:
if not os.path.isfile(input_file):
logger.error("Path should be file")
exit(1)
input_feeder = InputFeeder("video", input_file)
face_detector = FaceDetector(
args.face_detection_model,
device=args.device,
threshold=args.threshold,
extensions=args.extensions,
)
face_landmark_detector = FaceLandmarkDetector(
args.face_landmark_model,
device=args.device,
threshold=args.threshold,
extensions=args.extensions,
)
head_pose_estimator = HeadPoseEstimator(
args.head_pose_model,
device=args.device,
threshold=args.threshold,
extensions=args.extensions,
)
gaze_estimator = GazeEstimator(
args.gaze_estimation_model,
device=args.device,
threshold=args.threshold,
extensions=args.extensions,
)
mouse_controller = MouseController("medium", "fast")
face_detector.load_model()
face_landmark_detector.load_model()
head_pose_estimator.load_model()
gaze_estimator.load_model()
input_feeder.load_data()
width = 1000
height = int(width * 9 / 16)
for flag, frame in input_feeder.next_batch():
if not flag:
break
pressed_key = cv2.waitKey(60)
face_detected = face_detector.predict(frame)
if face_detected:
face_coordinates, face_image = face_detected
if not face_coordinates:
continue
else:
continue
if "fd" in args.visualization:
cv2.rectangle(
frame,
(face_coordinates[0], face_coordinates[1]),
(face_coordinates[2], face_coordinates[3]),
(36, 255, 12),
2,
)
cv2.putText(
frame,
"Face Detected",
(face_coordinates[0], face_coordinates[1] - 10),
cv2.FONT_HERSHEY_SIMPLEX,
0.9,
(36, 255, 12),
2,
)
left_eye_img, righ_eye_img, eye_coords = face_landmark_detector.predict(
face_image
)
if "fl" in args.visualization:
frame_eye_coords_min = (
np.array(eye_coords)[:, :2] + np.array(face_coordinates)[:2]
)
frame_eye_coords_max = (
np.array(eye_coords)[:, 2:] + np.array(face_coordinates)[:2]
)
cv2.rectangle(
frame,
(frame_eye_coords_min[0][0], frame_eye_coords_min[0][1]),
(frame_eye_coords_max[0][0], frame_eye_coords_max[0][1]),
(36, 255, 12),
2,
)
cv2.rectangle(
frame,
(frame_eye_coords_min[1][0], frame_eye_coords_min[1][1]),
(frame_eye_coords_max[1][0], frame_eye_coords_max[1][1]),
(36, 255, 12),
2,
)
head_pose_estimate = head_pose_estimator.predict(face_image)
if "hp" in args.visualization:
cv2.putText(
frame,
"yaw:{:.1f}|pitch:{:.1f}|roll:{:.1f}".format(*head_pose_estimate),
(20, 35),
cv2.FONT_HERSHEY_COMPLEX,
1.2,
(36, 255, 12),
3,
)
mouse_coordinate, gaze_vector = gaze_estimator.predict(
left_eye_img, righ_eye_img, head_pose_estimate
)
if "ge" in args.visualization:
head_pose_estimate = np.array(head_pose_estimate)
yaw, pitch, roll = head_pose_estimate * np.pi / 180.0
focal_length = 950
scale = 100
origin = (
int(
face_coordinates[0]
+ (face_coordinates[2] - face_coordinates[0]) / 2
),
int(
face_coordinates[1]
+ (face_coordinates[3] - face_coordinates[1]) / 2
),
)
r_x = np.array(
[
[1, 0, 0],
[0, math.cos(pitch), -math.sin(pitch)],
[0, math.sin(pitch), math.cos(pitch)],
]
)
r_y = np.array(
[
[math.cos(yaw), 0, -math.sin(yaw)],
[0, 1, 0],
[math.sin(yaw), 0, math.cos(yaw)],
]
)
r_z = np.array(
[
[math.cos(roll), -math.sin(roll), 0],
[math.sin(roll), math.cos(roll), 0],
[0, 0, 1],
]
)
r = r_z @ r_y @ r_x
zaxis = np.array(([0, 0, -1 * scale]), dtype="float32")
offset = np.array(([0, 0, focal_length]), dtype="float32")
zaxis = np.dot(r, zaxis) + offset
tip = (
int(zaxis[0] / zaxis[2] * focal_length) + origin[0],
int(zaxis[1] / zaxis[2] * focal_length) + origin[1],
)
cv2.arrowedLine(frame, origin, tip, (0, 0, 255), 3, tipLength=0.3)
cv2.imshow("frame", cv2.resize(frame, (width, height)))
mouse_controller.move(mouse_coordinate[0], mouse_coordinate[1])
if pressed_key == 27:
logger.error("exit key is pressed..")
break
import os
from io import BytesIO
from urllib.parse import urlencode
import asks
import cv2
from sanic import Sanic, response
from pydub import AudioSegment
from config import (BOT_TOKEN, HOST, PORT, DEBUG, DIRS)
from face_detection import FaceDetector
app = Sanic(__name__)
face_detector = FaceDetector()
API_URL = 'https://api.telegram.org/'
BOT_URL = API_URL + 'bot' + BOT_TOKEN + '/'
BOT_FILE_URL = API_URL + 'file/' + 'bot' + BOT_TOKEN
@app.post('/')
async def request_handler(request):
"""
Route function to handle request data depending on data's keys.
Return empty request with status code 204.
"""
message = request.json.get('message')
if 'photo' in message:
await photo_handler(message)
elif 'voice' in message:
await voice_handler(message)
def main():
"""Main entrypoint when running this module from the terminal."""
parser = argparse.ArgumentParser(
description='Process and a clean a raw dataset of halloween costumes')
parser.add_argument(
'dataset_source',
help='The path to the directory containing the source dataset.',
type=Path,
action=ReadableDirectory)
parser.add_argument(
'--destination',
'-d',
dest='dataset_destination',
help='The path to the directory which the cleaned '
'dataset should be saved. If this is not specified, the cleaned files are saved in the same parent '
'folder as the source.',
type=Path,
default=None)
parser.add_argument(
'--file-glob-patterns',
nargs='+',
type=str,
default=['*.png', '*.jpeg', '*.jpg'],
help='The glob patterns to use to find files in the source directory.')
parser.add_argument(
'--no-remove-transparency',
action='store_false',
dest='remove_transparency',
help='Remove transparency and replace it with a colour.')
parser.add_argument('--bg-colour',
type=str,
default='WHITE',
help='The colour to replace transparency with.')
parser.add_argument(
'--u2net-size',
type=str,
default='large',
help=
'The size of the pretrained U-2-net model. Either \'large\' or \'small\'.'
)
parser.add_argument(
'--face_image_ratio_threshold',
type=float,
default=0.05,
help='The maximum face-to-image area ratio that is allowed.')
parser.add_argument('--crop-faces',
dest='crop_faces',
action='store_true',
help='Crop out faces.')
parser.add_argument('--yes', '-y', action='store_true', help='Yes to all.')
args = parser.parse_args()
# Create a destination path if none was provided.
if args.dataset_destination is None:
args.dataset_destination = args.dataset_source.parent / (
args.dataset_source.stem + '_cleaned')
if args.dataset_destination.exists() and any(
args.dataset_destination.iterdir()):
if not args.yes:
click.confirm(
f'The destination path (\'{args.dataset_destination.resolve()}\') '
'already exists! Would you like to continue? This will overwrite the directory.',
abort=True)
_rmtree(args.dataset_destination)
args.dataset_destination.mkdir(exist_ok=True, parents=True)
u2net = U2Net(pretrained_model_name=args.u2net_size)
face_detector = FaceDetector()
files = list(get_files(args.dataset_source, args.file_glob_patterns))
with tqdm.tqdm(files) as progress:
for file in progress:
progress.set_description(f'Processing {file.name}')
# Skip images that don't have a single face in them...
face_detection_results = face_detector.detect_faces(file)
if len(face_detection_results) != 1:
continue
segmentation_map = u2net.segment_image(file)
try:
# Remove background from image (using U2Net)
image = u2net.remove_background(file, segmentation_map)
old_image_width, old_image_height = image.size
except InvalidImageError as e:
continue
# Crop image to bounding box (using U2Net)
bounding_box = u2net.get_bounding_box(segmentation_map)
image = image.crop(bounding_box)
# Tuple of the form (x1, y1, x2, y2)
fbb = face_detection_results[0].bounding_box
fbb_width = (fbb[2] - fbb[0])
fbb_height = (fbb[3] - fbb[1])
image_width, image_height = image.size
# Compute the face-to-image area ratio.
# This is used as a heuristic to filter out portrait images
# (i.e. when the face takes up more than a certain percentage of the total image).
face_image_ratio = (fbb_width * fbb_height) / (image_width *
image_height)
if face_image_ratio > args.face_image_ratio_threshold:
continue
if args.crop_faces:
# Crop out the face...
# This assumes that the image is of a person standing vertically.
# Convert the bottom-right y-coordinate of the face bounding box
# into the coordinate system AFTER cropping.
adjusted_fbb_y2 = fbb[3] - bounding_box[1]
image = image.crop((0, adjusted_fbb_y2 - fbb_height * 0.10,
image_width, image_height))
if args.remove_transparency:
# Replace transparency with colour
background_image = Image.new('RGBA', image.size,
args.bg_colour)
background_image.paste(image, (0, 0), image)
image = background_image.convert('RGB')
# Output processed image
destination = args.dataset_destination / (file.stem + '.png')
image.save(str(destination))
def main():
args = get_args()
log.basicConfig(filename='example.log', level=log.DEBUG)
inputFile = args.input
#inputFile = "./bin/demo.mp4"
mouse = MouseController("high", "fast")
frame_count = 0
focal_length = 950.0
scale = 50
#print(f"Visual flag: {args.visual_flag}")
if inputFile.lower() == "cam":
feed = InputFeeder('cam')
log.info("Video source: " + str(inputFile))
else:
if not os.path.isfile(inputFile):
log.error("Unable to find file: " + inputFile)
exit(1)
feed = InputFeeder("video", inputFile)
log.info("Video source: " + str(inputFile))
log.info("InputFeeder initialized")
log.info("Device: " + str(args.device))
log.info("Face detection model: " + str(args.facedetectionmodel))
log.info("Facial landmarks model: " + str(args.faciallandmarksmodel))
log.info("Head pose estimation model: " + str(args.headposemodel))
log.info("Gaze estimation model: " + str(args.gazeestimationmodel))
if args.stats == 1:
print("Running statistics...")
inference_times = []
fdm_inference_times = []
hpm_inference_times = []
flm_inference_times = []
gem_inference_times = []
start_time = time.time()
# Create instances of the different models
fdm = FaceDetector(args.facedetectionmodel, args.device,
args.cpu_extension)
if args.stats == 1:
start_time = time.time()
fdm.load_model()
fdm_load_time = time.time() - start_time
else:
fdm.load_model()
fdm.check_model()
hpm = HeadPoseEstimator(args.headposemodel, args.device,
args.cpu_extension)
if args.stats == 1:
start_time = time.time()
hpm.load_model()
hpm_load_time = time.time() - start_time
else:
hpm.load_model()
hpm.check_model()
flm = FacialLandmarksDetector(args.faciallandmarksmodel, args.device,
args.cpu_extension)
if args.stats == 1:
start_time = time.time()
flm.load_model()
flm_load_time = time.time() - start_time
else:
flm.load_model()
flm.check_model()
gem = GazeEstimator(args.gazeestimationmodel, args.device,
args.cpu_extension)
if args.stats == 1:
start_time = time.time()
gem.load_model()
gem_load_time = time.time() - start_time
else:
gem.load_model()
gem.check_model()
if args.stats == 1:
duration_loading = time.time() - start_time
print(
f"Duration for loading and checking the models: {duration_loading}"
)
log.info(
f"Duration for loading and checking the models: {duration_loading}"
)
cv2.namedWindow('preview', cv2.WINDOW_NORMAL)
cv2.resizeWindow('preview', 600, 600)
feed.load_data()
for ret, frame in feed.next_batch():
if not ret:
break
if frame is not None:
frame_count += 1
key = cv2.waitKey(60)
if args.stats == 1:
start_time = time.time()
# Run face detection
face_crop, face_coords = fdm.predict(frame.copy())
print("Face crop shape: " + str(face_crop.shape))
frame_h, frame_w = frame.shape[:2]
(xmin, ymin, xmax, ymax) = face_coords
face_frame = frame[ymin:ymax, xmin:xmax]
#center_of_face = (xmin + face_frame.shape[1] / 2, ymin + face_frame.shape[0] / 2, 0) # 0 for colour channel
#print("Center of face " + str(center_of_face))
try:
# Check if face was detected
if type(face_coords) == int:
print("Unable to detect face")
if key == 27:
break
continue
# Facial landmark detection
left_eye_crop, right_eye_crop, landmarks, crop_coords = flm.predict(
face_crop.copy())
#print("Landmarks" +str(landmarks))
left_eye = (landmarks[0], landmarks[1])
right_eye = (landmarks[2], landmarks[3])
# Landmark position based on complete frame
landmarks_viz = landmarks
landmarks_viz[0] = landmarks_viz[0] + xmin
landmarks_viz[1] = landmarks_viz[1] + ymin
landmarks_viz[2] = landmarks_viz[2] + xmin
landmarks_viz[3] = landmarks_viz[3] + ymin
crop_coords_viz = (crop_coords[0] + xmin, crop_coords[1] +
ymin, crop_coords[2] + xmin,
crop_coords[3] + ymin, crop_coords[4] +
xmin, crop_coords[5] + ymin,
crop_coords[6] + xmin,
crop_coords[7] + ymin)
left_eye_viz = (landmarks_viz[0], landmarks_viz[1])
right_eye_viz = (landmarks_viz[2], landmarks_viz[3])
third_eye_viz_x = (landmarks_viz[2] -
landmarks_viz[0]) / 2 + landmarks_viz[0]
third_eye_viz_y = (landmarks_viz[3] -
landmarks_viz[1]) / 2 + landmarks_viz[1]
third_eye_viz = (third_eye_viz_x, third_eye_viz_y)
#print(landmarks_viz[0], landmarks_viz[2], third_eye_viz_x)
# Head pose estimation
head_pose = hpm.predict(face_crop.copy())
print("Head pose: " + str(head_pose))
(yaw, pitch, roll) = head_pose
frame = display_head_pose(frame, pitch, roll, yaw)
# Send inputs to GazeEstimator
gaze_vector = gem.predict(head_pose, left_eye_crop,
right_eye_crop)
if args.stats == 1:
inference_time = time.time() - start_time
inference_times.append(inference_time)
print(gaze_vector)
frame = display_gaze(frame, gaze_vector)
# Control the mouse
if frame_count % 5 == 0:
mouse_x, mouse_y = get_mouse_vector(gaze_vector, roll)
print("Mouse vector:" + str(mouse_x) + " - " +
str(mouse_y))
mouse.move(mouse_x, mouse_y)
currentMouseX, currentMouseY = pyautogui.position()
print("Mouse coordinates: " + str(currentMouseX) + ", " +
str(currentMouseY))
if args.visual_flag == 1:
frame = draw_bounding_box(frame, face_coords)
left_eye_frame = crop_coords_viz[0:4]
right_eye_frame = crop_coords_viz[4:]
frame = draw_bounding_box(frame, left_eye_frame)
frame = draw_bounding_box(frame, right_eye_frame)
frame = visualize_landmark(frame, left_eye_viz)
frame = visualize_landmark(frame,
right_eye_viz,
color=(0, 0, 255))
frame = visualize_gaze(frame, gaze_vector, landmarks_viz)
# visualize the axes of the HeadPoseEstimator results
#frame = hpm.draw_axes(frame.copy(), center_of_face, yaw, pitch, roll, scale, focal_length)
frame = hpm.draw_axes(frame.copy(), third_eye_viz, yaw,
pitch, roll, scale, focal_length)
#hdm.draw_axes(frame.copy(), center_of_face, yaw, pitch, roll, scale, focal_length)
cv2.imshow('preview', frame)
cv2.imshow('left eye', left_eye_crop)
cv2.imshow('right eye', right_eye_crop)
except Exception as e:
print("Unable to predict using model" + str(e) +
" for frame " + str(frame_count))
log.error("Unable to predict using model" + str(e) +
" for frame " + str(frame_count))
continue
if args.stats == 1:
avg_inference_time = sum(inference_times) / len(inference_times)
print("Average inference time: " + str(avg_inference_time))
log.info("Average inference time: " + str(avg_inference_time))
log.info("Load time for face detection model: " + str(fdm_load_time))
log.info("Load time for facial landmarks model: " + str(flm_load_time))
log.info("Load time for head pose detection model: " +
str(hpm_load_time))
log.info("Load time for gaze estimation model: " + str(gem_load_time))
cv2.destroyAllWindows()
feed.close()
def parse_arguments(argv):
parser = argparse.ArgumentParser()
parser.add_argument('--src_path', type=str, help='src people image path')
parser.add_argument(
'--dst_path',
type=str,
help=
'move to dst people image path. remove image if this param is none.')
return parser.parse_args(argv)
if __name__ == '__main__':
sys.argv = [
'face_filter.py',
'--src_path',
'/home/xiajun/res/face/GC-WebFace/raw',
'--dst_path',
'/home/xiajun/res/face/GC-WebFace-tmp',
]
args = parse_arguments(sys.argv[1:])
print(args)
src_path = args.src_path
dst_path = args.dst_path
fdetector = FaceDetector()
face_filter(src_path, dst_path)
print('debug')
def main():
# Get command line arguments
args = parser.parse_args()
device = args.device
cpu_extensions = args.extensions
threshold = args.threshold
gaze_estimation_precision = args.gaze_estimation_precision
head_pose_precision = args.head_pose_precision
face_detection_precision = args.face_detection_precision
landmarks_precision = args.landmarks_precision
input_feeder = InputFeeder(args)
control_mouse = MouseController(args)
gaze_model = 'models/intel/gaze-estimation-adas-0002/{}/gaze-estimation-adas-0002'.format(
gaze_estimation_precision)
face_detector_model = 'models/intel/face-detection-adas-binary-0001/{}/face-detection-adas-binary-0001'.format(
face_detection_precision)
facial_landmark_model = 'models/intel/landmarks-regression-retail-0009/{}/landmarks-regression-retail-0009'.format(
landmarks_precision)
head_pose_model = 'models/intel/head-pose-estimation-adas-0001/{}/head-pose-estimation-adas-0001'.format(
head_pose_precision)
# Initialize the models
face_detector = FaceDetector(face_detector_model, args)
facial_landmarks = FacialLandmarksDetector(
model_name=facial_landmark_model,
device=device,
extensions=cpu_extensions)
head_pose_estimation = HeadPoseEstimation(model_name=head_pose_model,
device=device,
extensions=cpu_extensions)
gaze_estimation = GazeEstimation(model_name=gaze_model,
device=device,
extensions=cpu_extensions)
# Load the models
start_time = time.time()
face_detector.load_model()
face_detector_loadtime = time.time() - start_time
start_time = time.time()
facial_landmarks.load_model()
facial_landmark_loadtime = time.time() - start_time
start_time = time.time()
head_pose_estimation.load_model()
head_pose_estimation_loadtime = time.time() - start_time
start_time = time.time()
gaze_estimation.load_model()
gaze_estimation_loadtime = time.time() - start_time
logging.info('FINISH LOADING MODELS')
try:
width, height = input_feeder.load_data()
except TypeError:
logging.error('Invalid file type.')
return
output_handler = OutputHandler(args)
output_handler.initalize_video_writer(width, height)
frame_count = 0
start_time = 0
capture = input_feeder.cap
inputs = args.input
if input_feeder.input_type == 'cam':
inputs = 0
else:
capture.open(inputs)
while capture.isOpened():
flag, frame = capture.read()
if start_time == 0:
start_time = time.time()
if inputs == 0 and time.time() - start_time >= 1:
gaze_estimate = run_inference(frame, face_detector,
facial_landmarks,
head_pose_estimation,
gaze_estimation, output_handler)
if gaze_estimate is None:
break
if gaze_estimate[0][0]:
x, y = gaze_estimate[0][:2]
control_mouse.move(x, y)
start_time = 0
frame_count += 1
elif not inputs == 0:
gaze_estimate = run_inference(frame, face_detector,
facial_landmarks,
head_pose_estimation,
gaze_estimation, output_handler)
if gaze_estimate is None:
break
if gaze_estimate[0][0] and time.time() - start_time >= 0.5:
x, y = gaze_estimate[0][:2]
control_mouse.move(x, y)
start_time = 0
frame_count += 1
input_feeder.close()
logging.info('TOTOAL FRAMES PROCESSED: {}'.format(frame_count))
logging.info('Time to load face detector model is {:.5f}'.format(
face_detector_loadtime))
logging.info('Time to load head pose estimation model is {:.5f}'.format(
head_pose_estimation_loadtime))
logging.info('Time to load facial landmarks model model is {:.5f}'.format(
facial_landmark_loadtime))
logging.info('Time to load gaze estimation model is {:.5f}'.format(
gaze_estimation_loadtime))
def main(args):
fd_infer_time, ld_infer_time, hpe_infer_time, ge_infer_time = 0 ,0 ,0 ,0
start = time.time()
face_detector = FaceDetector(args.model_fd, args.device_fd, args.ext_fd)
fd_load_time = time.time() - start
start = time.time()
landmarks_detector = LandmarksDetector(args.model_ld, args.device_ld, args.ext_ld)
ld_load_time = time.time() - start
start = time.time()
head_pose_estimator = HeadPoseEstimator(args.model_hpe, args.device_hpe, args.ext_hpe)
hpe_load_time = time.time() - start
start = time.time()
gaze_estimator = GazeEstimator(args.model_ge, args.device_ge, args.ext_ge)
ge_load_time = time.time() - start
log.info("Models Loading...")
log.info("Face detection load time :{:.4f}ms".format(fd_load_time))
log.info("Landmarks estimation load time :{:.4f}ms".format(ld_load_time))
log.info("Head pose estimation load time :{:.4f}ms".format(hpe_load_time))
log.info("Gaze estimation load time :{:.4f}ms".format(ge_load_time))
log.info('All Models loaded')
mouse_controller = MouseController('high', 'fast')
if args.input == 0:
input_feeder = InputFeeder('cam', args.input)
elif args.input.endswith('.jpg') or args.input.endswith('.bmp'):
input_feeder = InputFeeder('image', args.input)
else:
input_feeder = InputFeeder('video', args.input)
input_feeder.load_data()
init_w = input_feeder.init_w
init_h = input_feeder.init_h
counter = 0
for flag, frame in input_feeder.next_batch():
if not flag:
break
counter +=1
key = cv2.waitKey(60)
try:
start = time.time()
outputs = face_detector.predict(frame)
face = face_detector.preprocess_output(frame, outputs, init_w, init_h)
fd_infer_time += time.time() - start
start = time.time()
outputs = landmarks_detector.predict(face)
left_eye, right_eye, real_landmraks = landmarks_detector.preprocess_output(face, outputs)
ld_infer_time += time.time() - start
start = time.time()
outputs = head_pose_estimator.predict(face)
head_pose_angles = head_pose_estimator.preprocess_output(outputs)
hpe_infer_time += time.time() - start
start = time.time()
outputs = gaze_estimator.predict(left_eye, right_eye, head_pose_angles)
gaze = gaze_estimator.preprocess_output(outputs)
ge_infer_time += time.time() - start
log.info("Face detection time :{:.4f}ms".format(fd_infer_time/counter))
log.info("Landmarks estimation time :{:.4f}ms".format(ld_infer_time/counter))
log.info("Head pose estimation time :{:.4f}ms".format(hpe_infer_time/counter))
log.info("Gaze estimation time :{:.4f}ms".format(ge_infer_time/counter))
if args.input != 0:
drawer = Drawer(face, real_landmraks, head_pose_angles, gaze)
drawer.draw_landmarks(20)
drawer.draw_head_pose()
drawer.draw_gazes()
drawer.show()
roll_cos = math.cos(head_pose_angles[2] * math.pi/180)
roll_sin = math.sin(head_pose_angles[2] * math.pi/180)
mouse_x = gaze[0] * roll_cos + gaze[0] * roll_sin
mouse_y = gaze[1] * roll_cos + gaze[1] * roll_sin
mouse_controller.move(mouse_x, mouse_y)
except Exception as e:
log.error(e)
finally:
if key == 27:
break
input_feeder.close()
args = parser.parse_args()
dir_name = args.name
detection_type = args.detection_type
only_face = True if args.only_face == 'yes' else False
capture = cv2.VideoCapture(0)
num = 1
while True:
if dir_name is None:
print('You must pass your name as argument!')
subprocess.run(["python", "took_pictures.py", "-h"])
break
img_dir = f'images/{dir_name}/'
os.makedirs(img_dir, exist_ok=True)
detector = FaceDetector(detection_type)
# Capture frame-by-frame
val, frame = capture.read()
gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
try:
start_x, start_y, end_x, end_y = detector.get_coordinates(frame)[0]
roi_ = frame[start_y:end_y, start_x:end_x]
cv2.rectangle(frame, (start_x, start_y), (end_x, end_y),
(255, 255, 255), 2)
if only_face:
cv2.imwrite(f'{img_dir}/{num}.png', roi_)
else:
cv2.imwrite(f'{img_dir}/{num}.png', frame)
print(f'face no. {num}')
num += 1
def main(args):
print("Main script running...")
log_name = 'stats_' + args.device + '_' + args.hpe + args.fld + args.ge
if not os.path.exists('output'):
os.makedirs('output')
print(f"Logging to: output/{log_name}")
log = open('output/' + log_name, 'w+')
print("Initializing models...")
fd = FaceDetector(
model_name=
'models/intel/face-detection-adas-binary-0001/FP32-INT1/face-detection-adas-binary-0001',
device=args.device,
extensions=None)
fd.load_model()
if args.v: print(f"Face Detection Load Time: {fd.load_time}")
hpe = HeadPoseEstimator(
model_name=
f'models/intel/head-pose-estimation-adas-0001/{args.hpe}/head-pose-estimation-adas-0001',
device=args.device,
extensions=None)
hpe.load_model()
if args.v: print(f"Head Pose Estimation Load Time: {hpe.load_time}")
fld = FacialLandmarkDetector(
model_name=
f'models/intel/landmarks-regression-retail-0009/{args.fld}/landmarks-regression-retail-0009',
device=args.device,
extensions=None)
fld.load_model()
if args.v: print(f"Facial Landmarks Detection Load Time: {fld.load_time}")
ge = GazeEstimator(
model_name=
f'models/intel/gaze-estimation-adas-0002/{args.ge}/gaze-estimation-adas-0002',
device=args.device,
extensions=None)
ge.load_model()
if args.v: print(f"Gaze Estimation Load Time: {ge.load_time}")
image = False
print("Initializing source feed...")
feed = InputFeeder(input_type=args.input_type, input_file=args.input_file)
if args.input_type == 'image':
image = True
feed.load_data()
for batch in feed.next_batch():
if args.v:
print()
cv2.imshow('Batch', batch)
if image:
cv2.imwrite('output/Batch.png', batch)
coords, bounding_face = fd.predict(batch)
if not coords:
print("No face")
continue
if image: cv2.imwrite('output/Face.png', bounding_face)
box = coords[0]
face = bounding_face[box[1]:box[3], box[0]:box[2]]
if args.v:
print(f"Face Time: {fd.infer_time}")
log.write("FD_infer: " + str(fd.infer_time) + "\n")
if image:
cv2.imshow('Cropped Face', face)
# Landmark Detection
coords, landmark_detection, landmark_points = fld.predict(face)
if image: cv2.imwrite('output/Landmarks.png', landmark_detection)
if image: cv2.imshow('Landmark Detection', landmark_detection)
if args.v: print(f"Landmark Time: {fld.infer_time}")
log.write("FLD_infer: " + str(fld.infer_time) + "\n")
right_box, left_box = coords[0:2]
if args.v: print(f"Eye Coords: {coords}")
if left_box == None or right_box == None:
print("No eyes")
continue
left_eye = face[left_box[1]:left_box[3], left_box[0]:left_box[2]]
cv2.putText(face, 'L', (left_box[0], left_box[3]),
cv2.FONT_HERSHEY_COMPLEX, 1, (0, 255, 0), 2)
right_eye = face[right_box[1]:right_box[3], right_box[0]:right_box[2]]
cv2.putText(face, 'R', (right_box[0], right_box[3]),
cv2.FONT_HERSHEY_COMPLEX, 1, (0, 255, 0), 2)
if args.v:
print(f"Eye Shape: {left_eye.shape} :: {right_eye.shape}")
#Head Pose Estimation
head_yaw, head_pitch, head_roll = hpe.predict(face)
if args.v: print(f"Head Pose Time: {hpe.infer_time}")
log.write("HPE_infer: " + str(hpe.infer_time) + "\n")
head_angles = [head_yaw[0][0], head_pitch[0][0], head_roll[0][0]]
#Gaze Estimation
# expects pose as (yaw, pitch, and roll)
gaze = ge.predict(left_eye, right_eye, head_angles)
if args.v:
print(f"Gaze Time: {ge.infer_time}")
log.write("GE_infer: " + str(ge.infer_time) + "\n")
gaze_point = (int(gaze[0][0] * 50), int(gaze[0][1] * 50))
arrows = cv2.arrowedLine(face, landmark_points[0],
(landmark_points[0][0] + gaze_point[0],
landmark_points[0][1] - gaze_point[1]),
(0, 0, 255), 2)
arrows = cv2.arrowedLine(face, landmark_points[1],
(landmark_points[1][0] + gaze_point[0],
landmark_points[1][1] - gaze_point[1]),
(0, 0, 255), 2)
if image:
cv2.imwrite('output/Gaze.png', arrows)
if not image:
mouse = MouseController(precision='medium', speed='medium')
mouse.move(gaze[0][0], gaze[0][1])
if image:
cv2.imshow('Arrows', arrows)
if image:
log.write("FD_LoadTime: " + str(fd.load_time) + "\n")
log.write("FD_PreprocessTime: " + str(fd.preprocess_input_time) +
"\n")
log.write("FD_PostrocessTime: " + str(fd.preprocess_output_time) +
"\n")
log.write("FLD_LoadTime: " + str(fld.load_time) + "\n")
log.write("FLD_PreprocessTime: " + str(fld.preprocess_input_time) +
"\n")
log.write("FLD_PostprocessTime: " +
str(fld.preprocess_output_time) + "\n")
log.write("HPE_LoadTime: " + str(hpe.load_time) + "\n")
log.write("HPE_PreprocessTime: " + str(hpe.preprocess_input_time) +
"\n")
log.write("GE_LoadTime: " + str(ge.load_time) + "\n")
log.write("GE_PreprocessTime: " + str(ge.preprocess_input_time) +
"\n")
cv2.waitKey(0)
else:
if cv2.waitKey(15) & 0xFF == ord('q'):
break
feed.close()
log.close()
cv2.destroyAllWindows