mirror of
https://github.com/serengil/deepface.git
synced 2025-06-10 13:27:08 +00:00
Merge pull request #1031 from AndreaLanfranchi/al20240221
Small knob polishing and amend duplicated attempts to add undetectable images to pickle
This commit is contained in:
Sefik Ilkin Serengil
GitHub
commit
14bbc2f938
@ -210,6 +210,5 @@ def align_face(
|
||||
return img, 0
|
||||
|
||||
angle = float(np.degrees(np.arctan2(right_eye[1] - left_eye[1], right_eye[0] - left_eye[0])))
|
||||
img = Image.fromarray(img)
|
||||
img = np.array(img.rotate(angle))
|
||||
img = np.array(Image.fromarray(img).rotate(angle))
|
||||
return img, angle
|
||||
|
||||
@ -40,11 +40,10 @@ def build_model(model_name: str) -> Any:
|
||||
if not "model_obj" in globals():
|
||||
model_obj = {}
|
||||
|
||||
if not model_name in model_obj:
|
||||
if not model_name in model_obj.keys():
|
||||
model = models.get(model_name)
|
||||
if model:
|
||||
model = model()
|
||||
model_obj[model_name] = model
|
||||
model_obj[model_name] = model()
|
||||
else:
|
||||
raise ValueError(f"Invalid model_name passed - {model_name}")
|
||||
|
||||
|
||||
@ -68,9 +68,8 @@ def analysis(
|
||||
|
||||
cap = cv2.VideoCapture(source) # webcam
|
||||
while True:
|
||||
_, img = cap.read()
|
||||
|
||||
if img is None:
|
||||
has_frame, img = cap.read()
|
||||
if not has_frame:
|
||||
break
|
||||
|
||||
# cv2.namedWindow('img', cv2.WINDOW_FREERATIO)
|
||||
@ -92,6 +91,8 @@ def analysis(
|
||||
faces = []
|
||||
for face_obj in face_objs:
|
||||
facial_area = face_obj["facial_area"]
|
||||
if facial_area["w"] <= 130: # discard small detected faces
|
||||
continue
|
||||
faces.append(
|
||||
(
|
||||
facial_area["x"],
|
||||
@ -111,36 +112,32 @@ def analysis(
|
||||
detected_faces = []
|
||||
face_index = 0
|
||||
for x, y, w, h in faces:
|
||||
if w > 130: # discard small detected faces
|
||||
face_detected = True
|
||||
if face_index == 0:
|
||||
face_included_frames += 1 # increase frame for a single face
|
||||
|
||||
face_detected = True
|
||||
if face_index == 0:
|
||||
face_included_frames = (
|
||||
face_included_frames + 1
|
||||
) # increase frame for a single face
|
||||
cv2.rectangle(
|
||||
img, (x, y), (x + w, y + h), (67, 67, 67), 1
|
||||
) # draw rectangle to main image
|
||||
|
||||
cv2.rectangle(
|
||||
img, (x, y), (x + w, y + h), (67, 67, 67), 1
|
||||
) # draw rectangle to main image
|
||||
cv2.putText(
|
||||
img,
|
||||
str(frame_threshold - face_included_frames),
|
||||
(int(x + w / 4), int(y + h / 1.5)),
|
||||
cv2.FONT_HERSHEY_SIMPLEX,
|
||||
4,
|
||||
(255, 255, 255),
|
||||
2,
|
||||
)
|
||||
|
||||
cv2.putText(
|
||||
img,
|
||||
str(frame_threshold - face_included_frames),
|
||||
(int(x + w / 4), int(y + h / 1.5)),
|
||||
cv2.FONT_HERSHEY_SIMPLEX,
|
||||
4,
|
||||
(255, 255, 255),
|
||||
2,
|
||||
)
|
||||
detected_face = img[int(y) : int(y + h), int(x) : int(x + w)] # crop detected face
|
||||
|
||||
detected_face = img[int(y) : int(y + h), int(x) : int(x + w)] # crop detected face
|
||||
# -------------------------------------
|
||||
|
||||
# -------------------------------------
|
||||
detected_faces.append((x, y, w, h))
|
||||
face_index = face_index + 1
|
||||
|
||||
detected_faces.append((x, y, w, h))
|
||||
face_index = face_index + 1
|
||||
|
||||
# -------------------------------------
|
||||
# -------------------------------------
|
||||
|
||||
if face_detected == True and face_included_frames == frame_threshold and freeze == False:
|
||||
freeze = True
|
||||
|
||||
@ -99,7 +99,7 @@ def find(
|
||||
|
||||
file_name = f"representations_{model_name}.pkl"
|
||||
file_name = file_name.replace("-", "_").lower()
|
||||
datastore_path = f"{db_path}/{file_name}"
|
||||
datastore_path = os.path.join(db_path, file_name)
|
||||
|
||||
df_cols = [
|
||||
"identity",
|
||||
@ -162,7 +162,7 @@ def find(
|
||||
logger.info(
|
||||
f"{len(newbies)} new representations are just added"
|
||||
f" whereas {len(oldies)} represented one(s) are just dropped"
|
||||
f" in {db_path}/{file_name} file."
|
||||
f" in {os.path.join(db_path,file_name)} file."
|
||||
)
|
||||
|
||||
if not silent:
|
||||
@ -173,8 +173,8 @@ def find(
|
||||
|
||||
if len(employees) == 0:
|
||||
raise ValueError(
|
||||
f"There is no image in {db_path} folder!"
|
||||
"Validate .jpg, .jpeg or .png files exist in this path.",
|
||||
f"Could not find any valid image in {db_path} folder!"
|
||||
"Valid images are .jpg, .jpeg or .png files.",
|
||||
)
|
||||
|
||||
# ------------------------
|
||||
@ -196,7 +196,7 @@ def find(
|
||||
pickle.dump(representations, f)
|
||||
|
||||
if not silent:
|
||||
logger.info(f"Representations stored in {db_path}/{file_name} file.")
|
||||
logger.info(f"Representations stored in {datastore_path} file.")
|
||||
|
||||
# ----------------------------
|
||||
# now, we got representations for facial database
|
||||
@ -241,6 +241,9 @@ def find(
|
||||
distances = []
|
||||
for _, instance in df.iterrows():
|
||||
source_representation = instance[f"{model_name}_representation"]
|
||||
if source_representation is None:
|
||||
distances.append(float("inf")) # no representation for this image
|
||||
continue
|
||||
|
||||
target_dims = len(list(target_representation))
|
||||
source_dims = len(list(source_representation))
|
||||
@ -292,7 +295,7 @@ def find(
|
||||
return resp_obj
|
||||
|
||||
|
||||
def __list_images(path: str) -> list:
|
||||
def __list_images(path: str) -> List[str]:
|
||||
"""
|
||||
List images in a given path
|
||||
Args:
|
||||
@ -304,7 +307,7 @@ def __list_images(path: str) -> list:
|
||||
for r, _, f in os.walk(path):
|
||||
for file in f:
|
||||
if file.lower().endswith((".jpg", ".jpeg", ".png")):
|
||||
exact_path = f"{r}/{file}"
|
||||
exact_path = os.path.join(r, file)
|
||||
images.append(exact_path)
|
||||
return images
|
||||
|
||||
@ -365,31 +368,35 @@ def __find_bulk_embeddings(
|
||||
expand_percentage=expand_percentage,
|
||||
)
|
||||
except ValueError as err:
|
||||
logger.warn(
|
||||
f"Exception while extracting faces from {employee}: {str(err)}. Skipping it."
|
||||
logger.error(
|
||||
f"Exception while extracting faces from {employee}: {str(err)}"
|
||||
)
|
||||
img_objs = []
|
||||
|
||||
for img_obj in img_objs:
|
||||
img_content = img_obj["face"]
|
||||
img_region = img_obj["facial_area"]
|
||||
embedding_obj = representation.represent(
|
||||
img_path=img_content,
|
||||
model_name=model_name,
|
||||
enforce_detection=enforce_detection,
|
||||
detector_backend="skip",
|
||||
align=align,
|
||||
normalization=normalization,
|
||||
)
|
||||
if len(img_objs) == 0:
|
||||
logger.warn(f"No face detected in {employee}. It will be skipped in detection.")
|
||||
representations.append((employee, None, 0, 0, 0, 0))
|
||||
else:
|
||||
for img_obj in img_objs:
|
||||
img_content = img_obj["face"]
|
||||
img_region = img_obj["facial_area"]
|
||||
embedding_obj = representation.represent(
|
||||
img_path=img_content,
|
||||
model_name=model_name,
|
||||
enforce_detection=enforce_detection,
|
||||
detector_backend="skip",
|
||||
align=align,
|
||||
normalization=normalization,
|
||||
)
|
||||
|
||||
img_representation = embedding_obj[0]["embedding"]
|
||||
img_representation = embedding_obj[0]["embedding"]
|
||||
representations.append((
|
||||
employee,
|
||||
img_representation,
|
||||
img_region["x"],
|
||||
img_region["y"],
|
||||
img_region["w"],
|
||||
img_region["h"]
|
||||
))
|
||||
|
||||
instance = []
|
||||
instance.append(employee)
|
||||
instance.append(img_representation)
|
||||
instance.append(img_region["x"])
|
||||
instance.append(img_region["y"])
|
||||
instance.append(img_region["w"])
|
||||
instance.append(img_region["h"])
|
||||
representations.append(instance)
|
||||
return representations
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user