github/deepface
1
1
Fork 0
mirror of https://github.com/serengil/deepface.git synced 2025-07-29 14:21:39 +00:00
Code Issues Packages Projects Releases Wiki Activity

Let preprocess_face always return region

Browse Source 
This makes the function easier to use, and lets use refactor later to a
list of regions.
This commit is contained in:
Roelof Ruis 2021-08-24 13:19:40 +02:00
parent 61e6de4048
commit 5adb041e19
4 changed files with 15 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
deepface/DeepFace.py
View File

@ -384,7 +384,7 @@ def analyze(img_path, actions = ['emotion', 'age', 'gender', 'race'] , models =
if action == 'emotion':
emotion_labels = ['angry', 'disgust', 'fear', 'happy', 'sad', 'surprise', 'neutral']
img, region = functions.preprocess_face(img = img_path, target_size = (48, 48), grayscale = True, enforce_detection = enforce_detection, detector_backend = detector_backend, return_region = True)
img, region = functions.preprocess_face(img = img_path, target_size = (48, 48), grayscale = True, enforce_detection = enforce_detection, detector_backend = detector_backend)
emotion_predictions = models['emotion'].predict(img)[0,:]
@ -401,7 +401,7 @@ def analyze(img_path, actions = ['emotion', 'age', 'gender', 'race'] , models =
elif action == 'age':
if img_224 is None:
img_224, region = functions.preprocess_face(img = img_path, target_size = (224, 224), grayscale = False, enforce_detection = enforce_detection, detector_backend = detector_backend, return_region = True)
img_224, region = functions.preprocess_face(img = img_path, target_size = (224, 224), grayscale = False, enforce_detection = enforce_detection, detector_backend = detector_backend)
age_predictions = models['age'].predict(img_224)[0,:]
apparent_age = Age.findApparentAge(age_predictions)
@ -410,7 +410,7 @@ def analyze(img_path, actions = ['emotion', 'age', 'gender', 'race'] , models =
elif action == 'gender':
if img_224 is None:
img_224, region = functions.preprocess_face(img = img_path, target_size = (224, 224), grayscale = False, enforce_detection = enforce_detection, detector_backend = detector_backend, return_region = True)
img_224, region = functions.preprocess_face(img = img_path, target_size = (224, 224), grayscale = False, enforce_detection = enforce_detection, detector_backend = detector_backend)
gender_prediction = models['gender'].predict(img_224)[0,:]
@ -423,7 +423,7 @@ def analyze(img_path, actions = ['emotion', 'age', 'gender', 'race'] , models =
elif action == 'race':
if img_224 is None:
img_224, region = functions.preprocess_face(img = img_path, target_size = (224, 224), grayscale = False, enforce_detection = enforce_detection, detector_backend = detector_backend, return_region = True) #just emotion model expects grayscale images
img_224, region = functions.preprocess_face(img = img_path, target_size = (224, 224), grayscale = False, enforce_detection = enforce_detection, detector_backend = detector_backend) #just emotion model expects grayscale images
race_predictions = models['race'].predict(img_224)[0,:]
race_labels = ['asian', 'indian', 'black', 'white', 'middle eastern', 'latino hispanic']
@ -745,7 +745,7 @@ def represent(img_path, model_name = 'VGG-Face', model = None, enforce_detection
input_shape_x, input_shape_y = functions.find_input_shape(model)
#detect and align
img = functions.preprocess_face(img = img_path
img, _ = functions.preprocess_face(img = img_path
, target_size=(input_shape_y, input_shape_x)
, enforce_detection = enforce_detection
, detector_backend = detector_backend
@ -810,8 +810,8 @@ def detectFace(img_path, detector_backend = 'opencv', enforce_detection = True,
deteced and aligned face in numpy format
"""
img = functions.preprocess_face(img = img_path, detector_backend = detector_backend
, enforce_detection = enforce_detection, align = align)[0] #preprocess_face returns (1, 224, 224, 3)
img, _ = functions.preprocess_face(img = img_path, detector_backend = detector_backend
, enforce_detection = enforce_detection, align = align) #preprocess_face returns (1, 224, 224, 3)
return img[:, :, ::-1] #bgr to rgb
#---------------------------

7
deepface/commons/functions.py
View File

@ -162,7 +162,7 @@ def normalize_input(img, normalization = 'base'):
return img
def preprocess_face(img, target_size=(224, 224), grayscale = False, enforce_detection = True, detector_backend = 'opencv', return_region = False, align = True):
def preprocess_face(img, target_size=(224, 224), grayscale = False, enforce_detection = True, detector_backend = 'opencv', align = True):
#img might be path, base64 or numpy array. Convert it to numpy whatever it is.
img = load_image(img)
@ -219,10 +219,7 @@ def preprocess_face(img, target_size=(224, 224), grayscale = False, enforce_dete
#---------------------------------------------------
if return_region == True:
return img_pixels, region
else:
return img_pixels
return img_pixels, region
def find_input_shape(model):

8
deepface/commons/realtime.py
View File

@ -96,7 +96,7 @@ def analysis(db_path, model_name = 'VGG-Face', detector_backend = 'opencv', dist
embedding = []
#preprocess_face returns single face. this is expected for source images in db.
img = functions.preprocess_face(img = employee, target_size = (input_shape_y, input_shape_x), enforce_detection = False, detector_backend = detector_backend)
img, _ = functions.preprocess_face(img = employee, target_size = (input_shape_y, input_shape_x), enforce_detection = False, detector_backend = detector_backend)
img_representation = model.predict(img)[0,:]
embedding.append(employee)
@ -202,7 +202,7 @@ def analysis(db_path, model_name = 'VGG-Face', detector_backend = 'opencv', dist
if enable_face_analysis == True:
gray_img = functions.preprocess_face(img = custom_face, target_size = (48, 48), grayscale = True, enforce_detection = False, detector_backend = 'opencv')
gray_img, _ = functions.preprocess_face(img = custom_face, target_size = (48, 48), grayscale = True, enforce_detection = False, detector_backend = 'opencv')
emotion_labels = ['Angry', 'Disgust', 'Fear', 'Happy', 'Sad', 'Surprise', 'Neutral']
emotion_predictions = emotion_model.predict(gray_img)[0,:]
sum_of_predictions = emotion_predictions.sum()
@ -280,7 +280,7 @@ def analysis(db_path, model_name = 'VGG-Face', detector_backend = 'opencv', dist
#-------------------------------
face_224 = functions.preprocess_face(img = custom_face, target_size = (224, 224), grayscale = False, enforce_detection = False, detector_backend = 'opencv')
face_224, _ = functions.preprocess_face(img = custom_face, target_size = (224, 224), grayscale = False, enforce_detection = False, detector_backend = 'opencv')
age_predictions = age_model.predict(face_224)[0,:]
apparent_age = Age.findApparentAge(age_predictions)
@ -335,7 +335,7 @@ def analysis(db_path, model_name = 'VGG-Face', detector_backend = 'opencv', dist
#-------------------------------
#face recognition
custom_face = functions.preprocess_face(img = custom_face, target_size = (input_shape_y, input_shape_x), enforce_detection = False, detector_backend = 'opencv')
custom_face, _ = functions.preprocess_face(img = custom_face, target_size = (input_shape_y, input_shape_x), enforce_detection = False, detector_backend = 'opencv')
#check preprocess_face function handled
if custom_face.shape[1:3] == input_shape:

4
tests/face-recognition-how.py
View File

@ -22,11 +22,11 @@ print("model output shape: ", model.layers[-1].input_shape[-1])
#load images and find embeddings
#img1 = functions.detectFace("dataset/img1.jpg", input_shape)
img1 = functions.preprocess_face("dataset/img1.jpg", input_shape)
img1, _ = functions.preprocess_face("dataset/img1.jpg", input_shape)
img1_representation = model.predict(img1)[0,:]
#img2 = functions.detectFace("dataset/img3.jpg", input_shape)
img2 = functions.preprocess_face("dataset/img3.jpg", input_shape)
img2, _ = functions.preprocess_face("dataset/img3.jpg", input_shape)
img2_representation = model.predict(img2)[0,:]
#----------------------------------------------