uvos/SDImagePreprocess
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

PersonDatasetAssembler: add the option to mach images that do NOT contain the specified person

Browse Source
This commit is contained in:
uvos
2024-06-07 14:04:07 +02:00
parent 03e2b3119a
commit f97f4640a9
1 changed files with 8 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
PersonDatasetAssembler/PersonDatasetAssembler.py
View File

@ -48,10 +48,11 @@ def extract_video_images(video: cv2.VideoCapture, interval: int = 0):
ret = True ret = True
frame_counter = 0 frame_counter = 0
while ret: while ret:
video.set(cv2.CAP_PROP_POS_FRAMES, frame_counter)
ret, frame = video.read() ret, frame = video.read()
if ret and frame_counter % interval == 0: if ret:
yield frame yield frame
frame_counter += 1 frame_counter += interval
def contains_face_match(detector: cv2.FaceDetectorYN, recognizer: cv2.FaceRecognizerSF, image: numpy.ndarray, referance_features: list(), thresh: float) -> bool: def contains_face_match(detector: cv2.FaceDetectorYN, recognizer: cv2.FaceRecognizerSF, image: numpy.ndarray, referance_features: list(), thresh: float) -> bool:
@ -88,6 +89,9 @@ def process_referance(detector: cv2.FaceDetectorYN, recognizer: cv2.FaceRecogniz
for image in images: for image in images:
detector.setInputSize([image.shape[1], image.shape[0]]) detector.setInputSize([image.shape[1], image.shape[0]])
faces = detector.detect(image)[1] faces = detector.detect(image)[1]
if faces is None:
print("unable to find face in referance image")
exit(1)
image = recognizer.alignCrop(image, faces[0]) image = recognizer.alignCrop(image, faces[0])
features = recognizer.feature(image) features = recognizer.feature(image)
out.append(features) out.append(features)
@ -103,6 +107,7 @@ if __name__ == "__main__":
parser.add_argument('--match_model', '-m', required=True, help="Path to the onnx recognition model to be used") parser.add_argument('--match_model', '-m', required=True, help="Path to the onnx recognition model to be used")
parser.add_argument('--detect_model', '-d', required=True, help="Path to the onnx detection model to be used") parser.add_argument('--detect_model', '-d', required=True, help="Path to the onnx detection model to be used")
parser.add_argument('--threshold', '-t', default=0.362, type=float, help="match threshold to use") parser.add_argument('--threshold', '-t', default=0.362, type=float, help="match threshold to use")
parser.add_argument('--invert', '-n', action='store_true', help="output files that DONT match")
args = parser.parse_args() args = parser.parse_args()
recognizer = cv2.FaceRecognizerSF.create(model=args.match_model, config="", backend_id=cv2.dnn.DNN_BACKEND_DEFAULT , target_id=cv2.dnn.DNN_TARGET_CPU) recognizer = cv2.FaceRecognizerSF.create(model=args.match_model, config="", backend_id=cv2.dnn.DNN_BACKEND_DEFAULT , target_id=cv2.dnn.DNN_TARGET_CPU)
@ -140,7 +145,7 @@ if __name__ == "__main__":
else: else:
resized = image resized = image
score, match = contains_face_match(detector, recognizer, resized, referance_features, args.threshold) score, match = contains_face_match(detector, recognizer, resized, referance_features, args.threshold)
if match: if match and not args.invert or not match and args.invert:
filename = f"{counter:04}.png" filename = f"{counter:04}.png"
cv2.imwrite(os.path.join(args.out, filename), image) cv2.imwrite(os.path.join(args.out, filename), image)
counter += 1 counter += 1