Spaces:
Running
Running
File size: 5,062 Bytes
cf27ba5 e425192 cf27ba5 |
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 |
import os
import cv2
import dlib
import mediapipe as mp
def read_image(image_path):
"""
Read an image from the given path and convert it to RGB.
"""
image = cv2.imread(image_path)
if image is None:
raise FileNotFoundError(f"Image not found at path: {image_path}")
return cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
class LandmarksDetector:
def __init__(self, predictor_model_path=f'{os.path.dirname(os.path.abspath(__file__))}/utils/shape_predictor_68_face_landmarks.dat'):
"""
:param predictor_model_path: path to shape_predictor_68_face_landmarks.dat file
"""
self.detector = dlib.get_frontal_face_detector() # cnn_face_detection_model_v1 also can be used
self.shape_predictor = dlib.shape_predictor(predictor_model_path)
def get_landmarks(self, image):
img = dlib.load_rgb_image(image)
dets = self.detector(img, 1)
dets = [dets[0]]
for detection in dets:
try:
face_landmarks = [(item.x, item.y) for item in self.shape_predictor(img, detection).parts()]
yield face_landmarks
except:
print("Exception in get_landmarks()!")
class DlibLandmarkDetector:
def __init__(self, predictor_model_path=f'{os.path.dirname(os.path.abspath(__file__))}/utils/shape_predictor_68_face_landmarks.dat'):
"""
:param predictor_model_path: path to shape_predictor_68_face_landmarks.dat file
"""
self.detector = dlib.get_frontal_face_detector() # cnn_face_detection_model_v1 also can be used
self.predictor = dlib.shape_predictor(predictor_model_path)
def get_landmarks(self, image_path):
# image = dlib.load_rgb_image(image_path)
image = read_image(image_path)
height, width, _ = image.shape
# Detect the faces in the image
dets = self.detector(image, 1) # 1 indicates to upsample the image 1 time. Higher values may give better results
# Raise an exception if no face is detected
if len(dets) == 0:
raise Exception("No face detected in the image at path: ", image_path)
# Get the landmarks of the first face detected
face_landmarks = [(item.x, item.y) for item in self.predictor(image, dets[0]).parts()]
# Add corner and edge midpoints as landmarks to include the background
corner_landmarks = [(1, 1), (1, height - 1), (width - 1, 1), (width - 1, height - 1)]
edge_landmarks = [(1, (height - 1)//2), ((width - 1)//2, 1), ((width - 1)//2, height - 1), (width - 1, (height - 1)//2)]
# Concatenate the landmarks
landmarks = face_landmarks + corner_landmarks + edge_landmarks
return landmarks, image
def show_landmarked_image(self, image_path, landmarks):
image = read_image(image_path)
for landmark in landmarks:
x, y = landmark
cv2.circle(image, (x, y), 1, (255, 255, 0), -1) # image, (x, y), radius, color, thickness (-1 to fill)
cv2.imshow('image', image)
cv2.waitKey(0)
cv2.destroyAllWindows()
class MediaPipeLandmarkDetector:
def __init__(self):
self.face_mesh = mp.solutions.face_mesh.FaceMesh(
static_image_mode=True,
max_num_faces=1,
min_detection_confidence=0.5)
def get_landmarks(self, image_path):
image = read_image(image_path)
height, width, _ = image.shape
image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# Process the image
results = self.face_mesh.process(image_rgb)
# Raise an exception if no face is detected
if results.multi_face_landmarks is None:
raise Exception("No face detected in the image at path: ", image_path)
# Extract the face landmarks
face_landmarks = results.multi_face_landmarks[0]
face_landmarks_normalized = [[landmark.x , landmark.y] for landmark in face_landmarks.landmark]
# Add corner and edge midpoints as landmarks to include the background
corner_landmarks = [(0, 0), (0, 1), (1, 0), (1, 1)]
edge_landmarks = [(0, 0.5), (0.5, 0), (0.5, 1), (1, 0.5)]
# Concatenate the corner and edge landmarks
landmarks = corner_landmarks + edge_landmarks + face_landmarks_normalized
# Multiply the landmarks with the image dimensions
landmarks = [(int(x * width) - 1, int(y * height) - 1) for x, y in landmarks]
landmarks = [(max(1, x), max(1, y)) for x, y in landmarks]
return landmarks, image
def show_landmarked_image(self, image_path, landmarks):
image = cv2.imread(image_path)
for landmark in landmarks:
x, y = landmark
cv2.circle(image, (x, y), 1, (255, 255, 0), -1)
cv2.imshow('image', image)
cv2.waitKey(0)
cv2.destroyAllWindows() |