danvics/immich-automated-selfie-timelapse
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Add back resolution threshold. Improve eye openness detection

Browse source
This commit is contained in:
Arnaud_Cayrol 2025-04-12 16:37:47 +02:00
parent fd4ef490c7
commit dd55dcbf3c
1 changed files with 16 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

29
timelapse.py
View file

@ -249,7 +249,7 @@ def get_head_pose(landmarks, image):
return pitch, yaw, roll return pitch, yaw, roll
def detect_landmarks(image, face_data, max_landmark_size=300): def detect_landmarks(image, face_data, face_resolution_threshold):
""" """
Detects facial landmarks in the image, resizing if necessary for better detection. Detects facial landmarks in the image, resizing if necessary for better detection.
@ -260,7 +260,7 @@ def detect_landmarks(image, face_data, max_landmark_size=300):
Returns: Returns:
dict or None: Dictionary containing facial landmarks in numpy arrays if successful, dict or None: Dictionary containing facial landmarks in numpy arrays if successful,
None if detection failed. None if face resolution is too low.
""" """
# Get face rectangle from metadata with proper scaling # Get face rectangle from metadata with proper scaling
face_img_width = face_data.get("imageWidth") face_img_width = face_data.get("imageWidth")
@ -274,14 +274,18 @@ def detect_landmarks(image, face_data, max_landmark_size=300):
y2 = int(face_data.get("boundingBoxY2", 0) * scale_y) y2 = int(face_data.get("boundingBoxY2", 0) * scale_y)
w = x2 - x1 w = x2 - x1
h = y2 - y1 h = y2 - y1
if w < face_resolution_threshold or h < face_resolution_threshold:
return None
# Crop the face region from the image # Crop the face region from the image
face_crop = image.crop((x1, y1, x2, y2)) face_crop = image.crop((x1, y1, x2, y2))
# If the cropped face is too large, resize it for better landmark detection # If the cropped face is too large, resize it for better landmark detection
optimal_size = 256 # optimal size for landmark detection
scale_factor = 1.0 scale_factor = 1.0
if w > max_landmark_size or h > max_landmark_size: if w > optimal_size or h > optimal_size:
scale_factor = max_landmark_size / max(w, h) scale_factor = optimal_size / max(w, h)
new_w = int(w * scale_factor) new_w = int(w * scale_factor)
new_h = int(h * scale_factor) new_h = int(h * scale_factor)
face_crop = face_crop.resize((new_w, new_h), Image.Resampling.LANCZOS) face_crop = face_crop.resize((new_w, new_h), Image.Resampling.LANCZOS)
@ -331,7 +335,7 @@ def detect_landmarks(image, face_data, max_landmark_size=300):
} }
def check_eye_visibility(left_eye, right_eye, ear_threshold=0.2): def check_eye_visibility(left_eye, right_eye, ear_threshold=0.2) -> bool:
""" """
Checks if both eyes are visible by calculating the Eye Aspect Ratio (EAR). Checks if both eyes are visible by calculating the Eye Aspect Ratio (EAR).
@ -341,7 +345,7 @@ def check_eye_visibility(left_eye, right_eye, ear_threshold=0.2):
ear_threshold (float): Threshold for eye visibility. ear_threshold (float): Threshold for eye visibility.
Returns: Returns:
tuple: (left_ear, right_ear) if both eyes are visible, None otherwise. bool: True if both eyes are open enough, False otherwise.
""" """
def calculate_ear(eye): def calculate_ear(eye):
v1 = np.linalg.norm(eye[1] - eye[5]) v1 = np.linalg.norm(eye[1] - eye[5])
@ -354,10 +358,9 @@ def check_eye_visibility(left_eye, right_eye, ear_threshold=0.2):
right_ear = calculate_ear(right_eye) right_ear = calculate_ear(right_eye)
if left_ear < ear_threshold or right_ear < ear_threshold: if left_ear < ear_threshold or right_ear < ear_threshold:
logger.info(f"Face turned too much to the side (EAR: L={left_ear:.2f}, R={right_ear:.2f})") return False
return None
return left_ear, right_ear return True
def crop_and_align_face(image, face_data, resize_size, face_resolution_threshold, pose_threshold, left_eye_pos): def crop_and_align_face(image, face_data, resize_size, face_resolution_threshold, pose_threshold, left_eye_pos):
@ -376,7 +379,7 @@ def crop_and_align_face(image, face_data, resize_size, face_resolution_threshold
PIL.Image or None: The aligned face image if successful, None otherwise. PIL.Image or None: The aligned face image if successful, None otherwise.
""" """
try: try:
# Get face rectangle from metadata with proper scaling # DEBUG : remove after testing
face_img_width = face_data.get("imageWidth") face_img_width = face_data.get("imageWidth")
face_img_height = face_data.get("imageHeight") face_img_height = face_data.get("imageHeight")
img_width, img_height = image.size img_width, img_height = image.size
@ -394,14 +397,14 @@ def crop_and_align_face(image, face_data, resize_size, face_resolution_threshold
img_np = cv2.cvtColor(img_np, cv2.COLOR_RGB2BGR) img_np = cv2.cvtColor(img_np, cv2.COLOR_RGB2BGR)
# Detect landmarks in the face region # Detect landmarks in the face region
landmarks = detect_landmarks(image, face_data) landmarks = detect_landmarks(image, face_data, face_resolution_threshold)
if not landmarks: if not landmarks:
logger.info("No facial landmarks detected") logger.info("Face resolution is too low")
draw_landmarks(img_np, None, (x1, y1, x2, y2), "discarded/no_landmarks.jpg")
return None return None
# Check if both eyes are visible # Check if both eyes are visible
if not check_eye_visibility(landmarks['left_eye'], landmarks['right_eye']): if not check_eye_visibility(landmarks['left_eye'], landmarks['right_eye']):
logger.info("Eyes are too closed")
draw_landmarks(img_np, landmarks['all_landmarks'], (x1, y1, x2, y2), "discarded/side_face.jpg") draw_landmarks(img_np, landmarks['all_landmarks'], (x1, y1, x2, y2), "discarded/side_face.jpg")
return None return None