def find(img_path,
db_path,
model_name='VGG-Face',
distance_metric='cosine',
model=None,
enforce_detection=True,
detector_backend='mtcnn'):
"""
This function applies verification several times and find an identity in a database
Parameters:
img_path: exact image path, numpy array or based64 encoded image. If you are going to find several identities, then you should pass img_path as array instead of calling find function in a for loop. e.g. img_path = ["img1.jpg", "img2.jpg"]
db_path (string): You should store some .jpg files in a folder and pass the exact folder path to this.
model_name (string): VGG-Face, Facenet, OpenFace, DeepFace, DeepID, Dlib or Ensemble
distance_metric (string): cosine, euclidean, euclidean_l2
model: built deepface model. A face recognition models are built in every call of find function. You can pass pre-built models to speed the function up.
model = DeepFace.build_model('VGG-Face')
enforce_detection (boolean): The function throws exception if a face could not be detected. Set this to True if you don't want to get exception. This might be convenient for low resolution images.
detector_backend (string): set face detector backend as mtcnn, opencv, ssd or dlib
Returns:
This function returns pandas data frame. If a list of images is passed to img_path, then it will return list of pandas data frame.
"""
tic = time.time()
img_paths, bulkProcess = functions.initialize_input(img_path)
functions.initialize_detector(detector_backend=detector_backend)
#-------------------------------
if os.path.isdir(db_path) == True:
if model == None:
if model_name == 'Ensemble':
print("Ensemble learning enabled")
models = Boosting.loadModel()
else: #model is not ensemble
model = build_model(model_name)
models = {}
models[model_name] = model
else: #model != None
print("Already built model is passed")
if model_name == 'Ensemble':
Boosting.validate_model(model)
models = model.copy()
else:
models = {}
models[model_name] = model
#---------------------------------------
if model_name == 'Ensemble':
model_names = ['VGG-Face', 'Facenet', 'OpenFace', 'DeepFace']
metric_names = ['cosine', 'euclidean', 'euclidean_l2']
elif model_name != 'Ensemble':
model_names = []
metric_names = []
model_names.append(model_name)
metric_names.append(distance_metric)
#---------------------------------------
file_name = "representations_%s.pkl" % (model_name)
file_name = file_name.replace("-", "_").lower()
if path.exists(db_path + "/" + file_name):
# print("WARNING: Representations for images in ",db_path," folder were previously stored in ", file_name, ". If you added new instances after this file creation, then please delete this file and call find function again. It will create it again.")
f = open(db_path + '/' + file_name, 'rb')
representations = pickle.load(f)
print("There are ", len(representations),
" representations found in ", file_name)
else: #create representation.pkl from scratch
employees = []
for r, d, f in os.walk(
db_path): # r=root, d=directories, f = files
for file in f:
if ('.jpg' in file.lower()) or ('.png' in file.lower()):
exact_path = r + "/" + file
employees.append(exact_path)
if len(employees) == 0:
raise ValueError(
"There is no image in ", db_path,
" folder! Validate .jpg or .png files exist in this path.")
#------------------------
#find representations for db images
representations = []
pbar = tqdm(range(0, len(employees)),
desc='Finding representations')
#for employee in employees:
for index in pbar:
employee = employees[index]
instance = []
instance.append(employee)
for j in model_names:
custom_model = models[j]
#----------------------------------
#decide input shape
input_shape = functions.find_input_shape(custom_model)
input_shape_x = input_shape[0]
input_shape_y = input_shape[1]
#----------------------------------
img = functions.preprocess_face(
img=employee,
target_size=(input_shape_y, input_shape_x),
enforce_detection=enforce_detection,
detector_backend=detector_backend)
representation = custom_model.predict(img)[0, :]
instance.append(representation)
#-------------------------------
representations.append(instance)
f = open(db_path + '/' + file_name, "wb")
pickle.dump(representations, f)
f.close()
print(
"Representations stored in ", db_path, "/", file_name,
" file. Please delete this file when you add new identities in your database."
)
#----------------------------
#now, we got representations for facial database
if model_name != 'Ensemble':
df = pd.DataFrame(
representations,
columns=["identity",
"%s_representation" % (model_name)])
else: #ensemble learning
columns = ['identity']
[columns.append('%s_representation' % i) for i in model_names]
df = pd.DataFrame(representations, columns=columns)
df_base = df.copy(
) #df will be filtered in each img. we will restore it for the next item.
resp_obj = []
global_pbar = tqdm(range(0, len(img_paths)), desc='Analyzing')
for j in global_pbar:
img_path = img_paths[j]
#find representation for passed image
for j in model_names:
custom_model = models[j]
#--------------------------------
#decide input shape
input_shape = functions.find_input_shape(custom_model)
input_shape_x = input_shape[0]
input_shape_y = input_shape[1]
#--------------------------------
img = functions.preprocess_face(
img=img_path,
target_size=(input_shape_y, input_shape_x),
enforce_detection=enforce_detection,
detector_backend=detector_backend)
target_representation = custom_model.predict(img)[0, :]
for k in metric_names:
distances = []
for index, instance in df.iterrows():
source_representation = instance["%s_representation" %
(j)]
if k == 'cosine':
distance = dst.findCosineDistance(
source_representation, target_representation)
elif k == 'euclidean':
distance = dst.findEuclideanDistance(
source_representation, target_representation)
elif k == 'euclidean_l2':
distance = dst.findEuclideanDistance(
dst.l2_normalize(source_representation),
dst.l2_normalize(target_representation))
distances.append(distance)
#---------------------------
if model_name == 'Ensemble' and j == 'OpenFace' and k == 'euclidean':
continue
else:
df["%s_%s" % (j, k)] = distances
if model_name != 'Ensemble':
threshold = dst.findThreshold(j, k)
df = df.drop(columns=["%s_representation" % (j)])
df = df[df["%s_%s" % (j, k)] <= threshold]
df = df.sort_values(
by=["%s_%s" % (j, k)],
ascending=True).reset_index(drop=True)
resp_obj.append(df)
df = df_base.copy(
) #restore df for the next iteration
#----------------------------------
if model_name == 'Ensemble':
feature_names = []
for j in model_names:
for k in metric_names:
if model_name == 'Ensemble' and j == 'OpenFace' and k == 'euclidean':
continue
else:
feature = '%s_%s' % (j, k)
feature_names.append(feature)
#print(df.head())
x = df[feature_names].values
#--------------------------------------
boosted_tree = Boosting.build_gbm()
y = boosted_tree.predict(x)
verified_labels = []
scores = []
for i in y:
verified = np.argmax(i) == 1
score = i[np.argmax(i)]
verified_labels.append(verified)
scores.append(score)
df['verified'] = verified_labels
df['score'] = scores
df = df[df.verified == True]
#df = df[df.score > 0.99] #confidence score
df = df.sort_values(by=["score"],
ascending=False).reset_index(drop=True)
df = df[['identity', 'verified', 'score']]
resp_obj.append(df)
df = df_base.copy() #restore df for the next iteration
#----------------------------------
toc = time.time()
print("find function lasts ", toc - tic, " seconds")
if len(resp_obj) == 1:
return resp_obj[0]
return resp_obj
else:
raise ValueError("Passed db_path does not exist!")
return None
def verify(img1_path,
img2_path='',
model_name='VGG-Face',
distance_metric='cosine',
model=None,
enforce_detection=True,
detector_backend='mtcnn'):
tic = time.time()
img_list, bulkProcess = initialize_input(img1_path, img2_path)
functions.initialize_detector(detector_backend=detector_backend)
resp_objects = []
#--------------------------------
if model_name == 'Ensemble':
model_names = ["VGG-Face", "Facenet", "OpenFace", "DeepFace"]
metrics = ["cosine", "euclidean", "euclidean_l2"]
else:
model_names = []
metrics = []
model_names.append(model_name)
metrics.append(distance_metric)
#--------------------------------
if model == None:
if model_name == 'Ensemble':
models = Boosting.loadModel()
else:
model = build_model(model_name)
models = {}
models[model_name] = model
else:
if model_name == 'Ensemble':
Boosting.validate_model(model)
else:
models = {}
models[model_name] = model
#------------------------------
#calling deepface in a for loop causes lots of progress bars. this prevents it.
disable_option = False if len(img_list) > 1 else True
pbar = tqdm(range(0, len(img_list)),
desc='Verification',
disable=disable_option)
for index in pbar:
instance = img_list[index]
if type(instance) == list and len(instance) >= 2:
img1_path = instance[0]
img2_path = instance[1]
ensemble_features = []
for i in model_names:
custom_model = models[i]
#decide input shape
input_shape = functions.find_input_shape(custom_model)
input_shape_x = input_shape[0]
input_shape_y = input_shape[1]
#----------------------
#detect and align faces
img1 = functions.preprocess_face(
img=img1_path,
target_size=(input_shape_y, input_shape_x),
enforce_detection=enforce_detection,
detector_backend=detector_backend)
img2 = functions.preprocess_face(
img=img2_path,
target_size=(input_shape_y, input_shape_x),
enforce_detection=enforce_detection,
detector_backend=detector_backend)
#----------------------
#find embeddings
img1_representation = custom_model.predict(img1)[0, :]
img2_representation = custom_model.predict(img2)[0, :]
#----------------------
#find distances between embeddings
for j in metrics:
if j == 'cosine':
distance = dst.findCosineDistance(
img1_representation, img2_representation)
elif j == 'euclidean':
distance = dst.findEuclideanDistance(
img1_representation, img2_representation)
elif j == 'euclidean_l2':
distance = dst.findEuclideanDistance(
dst.l2_normalize(img1_representation),
dst.l2_normalize(img2_representation))
else:
raise ValueError("Invalid distance_metric passed - ",
distance_metric)
#----------------------
#decision
if model_name != 'Ensemble':
threshold = dst.findThreshold(i, j)
if distance <= threshold:
identified = True
else:
identified = False
resp_obj = {
"verified": identified,
"distance": distance,
"max_threshold_to_verify": threshold,
"model": model_name,
"similarity_metric": distance_metric
}
if bulkProcess == True:
resp_objects.append(resp_obj)
else:
return resp_obj
else: #Ensemble
#this returns same with OpenFace - euclidean_l2
if i == 'OpenFace' and j == 'euclidean':
continue
else:
ensemble_features.append(distance)
#----------------------
if model_name == 'Ensemble':
boosted_tree = Boosting.build_gbm()
prediction = boosted_tree.predict(
np.expand_dims(np.array(ensemble_features), axis=0))[0]
verified = np.argmax(prediction) == 1
score = prediction[np.argmax(prediction)]
resp_obj = {
"verified": verified,
"score": score,
"distance": ensemble_features,
"model": ["VGG-Face", "Facenet", "OpenFace", "DeepFace"],
"similarity_metric":
["cosine", "euclidean", "euclidean_l2"]
}
if bulkProcess == True:
resp_objects.append(resp_obj)
else:
return resp_obj
#----------------------
else:
raise ValueError("Invalid arguments passed to verify function: ",
instance)
#-------------------------
toc = time.time()
if bulkProcess == True:
resp_obj = {}
for i in range(0, len(resp_objects)):
resp_item = resp_objects[i]
resp_obj["pair_%d" % (i + 1)] = resp_item
return resp_obj
def verify(img1_path,
img2_path='',
model_name='VGG-Face',
distance_metric='cosine',
model=None,
enforce_detection=True,
detector_backend='mtcnn'):
"""
This function verifies an image pair is same person or different persons.
Parameters:
img1_path, img2_path: exact image path, numpy array or based64 encoded images could be passed. If you are going to call verify function for a list of image pairs, then you should pass an array instead of calling the function in for loops.
e.g. img1_path = [
['img1.jpg', 'img2.jpg'],
['img2.jpg', 'img3.jpg']
]
model_name (string): VGG-Face, Facenet, OpenFace, DeepFace, DeepID, Dlib, ArcFace or Ensemble
distance_metric (string): cosine, euclidean, euclidean_l2
model: Built deepface model. A face recognition model is built every call of verify function. You can pass pre-built face recognition model optionally if you will call verify function several times.
model = DeepFace.build_model('VGG-Face')
enforce_detection (boolean): If any face could not be detected in an image, then verify function will return exception. Set this to False not to have this exception. This might be convenient for low resolution images.
detector_backend (string): set face detector backend as mtcnn, opencv, ssd or dlib
Returns:
Verify function returns a dictionary. If img1_path is a list of image pairs, then the function will return list of dictionary.
{
"verified": True
, "distance": 0.2563
, "max_threshold_to_verify": 0.40
, "model": "VGG-Face"
, "similarity_metric": "cosine"
}
"""
tic = time.time()
img_list, bulkProcess = functions.initialize_input(img1_path, img2_path)
functions.initialize_detector(detector_backend=detector_backend)
resp_objects = []
#--------------------------------
if model_name == 'Ensemble':
model_names = ["VGG-Face", "Facenet", "OpenFace", "DeepFace"]
metrics = ["cosine", "euclidean", "euclidean_l2"]
else:
model_names = []
metrics = []
model_names.append(model_name)
metrics.append(distance_metric)
#--------------------------------
if model == None:
if model_name == 'Ensemble':
models = Boosting.loadModel()
else:
model = build_model(model_name)
models = {}
models[model_name] = model
else:
if model_name == 'Ensemble':
Boosting.validate_model(model)
models = model.copy()
else:
models = {}
models[model_name] = model
#------------------------------
#calling deepface in a for loop causes lots of progress bars. this prevents it.
disable_option = False if len(img_list) > 1 else True
pbar = tqdm(range(0, len(img_list)),
desc='Verification',
disable=disable_option)
for index in pbar:
instance = img_list[index]
if type(instance) == list and len(instance) >= 2:
img1_path = instance[0]
img2_path = instance[1]
ensemble_features = []
for i in model_names:
custom_model = models[i]
#decide input shape
input_shape = functions.find_input_shape(custom_model)
input_shape_x = input_shape[0]
input_shape_y = input_shape[1]
#----------------------
#detect and align faces
img1 = functions.preprocess_face(
img=img1_path,
target_size=(input_shape_y, input_shape_x),
enforce_detection=enforce_detection,
detector_backend=detector_backend)
img2 = functions.preprocess_face(
img=img2_path,
target_size=(input_shape_y, input_shape_x),
enforce_detection=enforce_detection,
detector_backend=detector_backend)
#----------------------
#find embeddings
img1_representation = custom_model.predict(img1)[0, :]
img2_representation = custom_model.predict(img2)[0, :]
#----------------------
#find distances between embeddings
for j in metrics:
if j == 'cosine':
distance = dst.findCosineDistance(
img1_representation, img2_representation)
elif j == 'euclidean':
distance = dst.findEuclideanDistance(
img1_representation, img2_representation)
elif j == 'euclidean_l2':
distance = dst.findEuclideanDistance(
dst.l2_normalize(img1_representation),
dst.l2_normalize(img2_representation))
else:
raise ValueError("Invalid distance_metric passed - ",
distance_metric)
distance = np.float64(
distance
) #causes trobule for euclideans in api calls if this is not set (issue #175)
#----------------------
#decision
if model_name != 'Ensemble':
threshold = dst.findThreshold(i, j)
if distance <= threshold:
identified = True
else:
identified = False
resp_obj = {
"verified": identified,
"distance": distance,
"max_threshold_to_verify": threshold,
"model": model_name,
"similarity_metric": distance_metric
}
if bulkProcess == True:
resp_objects.append(resp_obj)
else:
return resp_obj
else: #Ensemble
#this returns same with OpenFace - euclidean_l2
if i == 'OpenFace' and j == 'euclidean':
continue
else:
ensemble_features.append(distance)
#----------------------
if model_name == 'Ensemble':
boosted_tree = Boosting.build_gbm()
prediction = boosted_tree.predict(
np.expand_dims(np.array(ensemble_features), axis=0))[0]
verified = np.argmax(prediction) == 1
score = prediction[np.argmax(prediction)]
resp_obj = {
"verified": verified,
"score": score,
"distance": ensemble_features,
"model": ["VGG-Face", "Facenet", "OpenFace", "DeepFace"],
"similarity_metric":
["cosine", "euclidean", "euclidean_l2"]
}
if bulkProcess == True:
resp_objects.append(resp_obj)
else:
return resp_obj
#----------------------
else:
raise ValueError("Invalid arguments passed to verify function: ",
instance)
#-------------------------
toc = time.time()
if bulkProcess == True:
resp_obj = {}
for i in range(0, len(resp_objects)):
resp_item = resp_objects[i]
resp_obj["pair_%d" % (i + 1)] = resp_item
return resp_obj
for i in range(0, len(dataset)):
item = resp_obj['pair_%s' % (i + 1)]
verified = item["verified"]
score = item["score"]
print(verified)
#-----------------------------------
print("--------------------------")
print("Pre-trained ensemble method - find")
from deepface import DeepFace
from deepface.basemodels import Boosting
model = Boosting.loadModel()
df = DeepFace.find("dataset/img1.jpg",
db_path="dataset",
model_name='Ensemble',
model=model,
enforce_detection=False)
print(df)
#-----------------------------------
print("--------------------------")
print("Pre-trained ensemble method - verify")
res = DeepFace.verify(dataset, model_name="Ensemble", model=model)
print(res)