Spaces:
Sleeping
Sleeping
File size: 1,732 Bytes
aae9c6b |
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 |
import cv2
import numpy as np
from tqdm import tqdm
def detect_faces(frames, min_face_size=30):
"""
Detect faces in frames using OpenCV's Haar Cascade classifier instead of dlib
Args:
frames: List of frames to detect faces in
min_face_size: Minimum face size to detect
Returns:
Tuple of (face_frames, count) where face_frames is a numpy array of detected faces
and count is the number of faces detected
"""
# Initialize face detector
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
# Initialize array to store faces
temp_face = np.zeros((len(frames), 224, 224, 3), dtype=np.uint8)
count = 0
# Process each frame
for _, frame in tqdm(enumerate(frames), total=len(frames)):
# Convert to grayscale for face detection
gray = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)
gray = cv2.cvtColor(gray, cv2.COLOR_BGR2GRAY)
# Detect faces
faces = face_cascade.detectMultiScale(
gray,
scaleFactor=1.1,
minNeighbors=5,
minSize=(min_face_size, min_face_size)
)
# Process each detected face
for (x, y, w, h) in faces:
if count < len(frames):
# Extract and resize face
face_image = frame[y:y+h, x:x+w]
face_image = cv2.resize(face_image, (224, 224), interpolation=cv2.INTER_AREA)
# Store face
temp_face[count] = face_image
count += 1
else:
break
return ([], 0) if count == 0 else (temp_face[:count], count) |