Scoring working!
Signed-off-by: James Ketrenos <james_git@ketrenos.com>
This commit is contained in:
parent
7a960c5f1f
commit
81e8a12546
@ -9,6 +9,10 @@ import numpy as np
|
||||
from deepface import DeepFace
|
||||
from retinaface import RetinaFace
|
||||
|
||||
|
||||
sqlite3.register_adapter(np.array, lambda arr: arr.tobytes())
|
||||
sqlite3.register_converter("array", np.frombuffer)
|
||||
|
||||
class NpEncoder(json.JSONEncoder):
|
||||
def default(self, obj):
|
||||
if isinstance(obj, np.integer):
|
||||
@ -96,6 +100,29 @@ def update_face_count(conn, photoId, faces):
|
||||
conn.commit()
|
||||
return None
|
||||
|
||||
def findCosineDistance(source_representation, test_representation):
|
||||
if type(source_representation) == list:
|
||||
source_representation = np.array(source_representation)
|
||||
if type(test_representation) == list:
|
||||
test_representation = np.array(test_representation)
|
||||
a = np.matmul(np.transpose(source_representation), test_representation)
|
||||
b = np.sum(np.multiply(source_representation, source_representation))
|
||||
c = np.sum(np.multiply(test_representation, test_representation))
|
||||
return 1 - (a / (np.sqrt(b) * np.sqrt(c)))
|
||||
|
||||
def findEuclideanDistance(source_representation, test_representation):
|
||||
if type(source_representation) == list:
|
||||
source_representation = np.array(source_representation)
|
||||
if type(test_representation) == list:
|
||||
test_representation = np.array(test_representation)
|
||||
euclidean_distance = source_representation - test_representation
|
||||
euclidean_distance = np.sum(np.multiply(euclidean_distance, euclidean_distance))
|
||||
euclidean_distance = np.sqrt(euclidean_distance)
|
||||
return euclidean_distance
|
||||
|
||||
def l2_normalize(x):
|
||||
return x / np.sqrt(np.sum(np.multiply(x, x)))
|
||||
|
||||
base = '/pictures/'
|
||||
conn = create_connection('../db/photos.db')
|
||||
faces = {}
|
||||
@ -106,16 +133,55 @@ with conn:
|
||||
res = cur.execute('''
|
||||
SELECT faces.id,facedescriptors.descriptors
|
||||
FROM faces
|
||||
LEFT JOIN facedescriptors ON (faces.descriptorId=facedescriptors.id)
|
||||
JOIN facedescriptors ON (faces.descriptorId=facedescriptors.id)
|
||||
WHERE faces.identityId IS null
|
||||
''')
|
||||
for row in res.fetchall():
|
||||
id, descriptors = row
|
||||
if faces[id] is None:
|
||||
face = {}
|
||||
faces[id] = face
|
||||
else:
|
||||
if id in faces:
|
||||
face = faces[id]
|
||||
face['descriptors'] = descriptors
|
||||
else:
|
||||
face = {
|
||||
'id': id,
|
||||
'scanned': False
|
||||
}
|
||||
faces[id] = face
|
||||
face['descriptors'] = np.frombuffer(descriptors)
|
||||
|
||||
for key1 in faces:
|
||||
face1 = faces[key1]
|
||||
if face1['scanned'] == True:
|
||||
continue
|
||||
face1['scanned'] = True
|
||||
for key2 in faces:
|
||||
if key1 == key2:
|
||||
continue
|
||||
face2 = faces[key2]
|
||||
face = {
|
||||
'between': (face1['id'], face2['id'])
|
||||
}
|
||||
|
||||
face['distanceCosine'] = findCosineDistance(
|
||||
face1['descriptors'],
|
||||
face2['descriptors']
|
||||
)
|
||||
face['distanceEuclidian'] = findEuclideanDistance(
|
||||
face1['descriptors'],
|
||||
face2['descriptors']
|
||||
)
|
||||
face['distanceEuclidianL2'] = findEuclideanDistance(
|
||||
l2_normalize(face1['descriptors']),
|
||||
l2_normalize(face2['descriptors'])
|
||||
)
|
||||
face['scoring'] = 0
|
||||
if face['distanceCosine'] >= 0.68:
|
||||
face['scoring'] += 1
|
||||
if face['distanceEuclidian'] >= 4.15:
|
||||
face['scoring'] += 1
|
||||
if face['distanceEuclidianL2'] >= 1.13:
|
||||
face['scoring'] += 1
|
||||
|
||||
if face['scoring'] == 3: # Same face!
|
||||
print(face)
|
||||
|
||||
# update_face_count(conn, photoId, len(faces))
|
||||
|
||||
@ -62,11 +62,10 @@ def extract_faces(img, threshold=0.75, model = None, allow_upscaling = True):
|
||||
# Re-implementation of 'extract_faces' with the addition of keeping a
|
||||
# copy of the face image for caching on disk
|
||||
if type(faces) == dict:
|
||||
print(f'Found {len(faces)} faces')
|
||||
i=1
|
||||
k=1
|
||||
for key in faces:
|
||||
print(f'Processing face {i}/{len(faces)}')
|
||||
i+=1
|
||||
print(f'Processing face {k}/{len(faces)}')
|
||||
k+=1
|
||||
identity = faces[key]
|
||||
facial_area = identity["facial_area"]
|
||||
landmarks = identity["landmarks"]
|
||||
@ -112,7 +111,7 @@ def extract_faces(img, threshold=0.75, model = None, allow_upscaling = True):
|
||||
# Eye order is reversed as the routine does them backwards
|
||||
image = Image.fromarray(facial_img)
|
||||
image = alignment_procedure(image, right_eye, left_eye)
|
||||
#image = image.resize(size = input_shape, resample = Image.LANCZOS)
|
||||
image = image.resize(size = input_shape, resample = Image.LANCZOS)
|
||||
resized = np.asarray(image)
|
||||
|
||||
identity['vector'] = DeepFace.represent(
|
||||
@ -122,7 +121,6 @@ def extract_faces(img, threshold=0.75, model = None, allow_upscaling = True):
|
||||
detector_backend = 'retinaface',
|
||||
enforce_detection = False)
|
||||
|
||||
print(img.shape)
|
||||
identity["face"] = {
|
||||
'top': facial_area[1] / img.shape[0],
|
||||
'left': facial_area[0] / img.shape[1],
|
||||
@ -159,8 +157,8 @@ def create_face(conn, face):
|
||||
:return: face id
|
||||
"""
|
||||
sql = '''
|
||||
INSERT INTO faces(photoId,scanVersion,faceConfidence,top,left,bottom,right)
|
||||
VALUES(?,?,?,?,?,?,?)
|
||||
INSERT INTO faces(photoId,scanVersion,faceConfidence,top,left,bottom,right,descriptorId)
|
||||
VALUES(?,?,?,?,?,?,?,?)
|
||||
'''
|
||||
cur = conn.cursor()
|
||||
cur.execute(sql, (
|
||||
@ -170,7 +168,8 @@ def create_face(conn, face):
|
||||
face['top'],
|
||||
face['left'],
|
||||
face['bottom'],
|
||||
face['right']
|
||||
face['right'],
|
||||
face['descriptorId']
|
||||
))
|
||||
conn.commit()
|
||||
return cur.lastrowid
|
||||
@ -179,8 +178,7 @@ def create_face_descriptor(conn, face):
|
||||
"""
|
||||
Create a new face in the faces table
|
||||
:param conn:
|
||||
:param faceId:
|
||||
:param descriptor:
|
||||
:param face:
|
||||
:return: descriptor id
|
||||
"""
|
||||
sql = '''
|
||||
@ -232,10 +230,12 @@ with conn:
|
||||
if faces is None:
|
||||
update_face_count(conn, photoId, 0)
|
||||
continue
|
||||
print(f'Handling {len(faces)} faces')
|
||||
j=1
|
||||
for key in faces:
|
||||
face = faces[key]
|
||||
image = face['image']
|
||||
print(f'Writing face {j}/{len(faces)}')
|
||||
j+=1
|
||||
|
||||
#face['analysis'] = DeepFace.analyze(img_path = img, actions = ['age', 'gender', 'race', 'emotion'], enforce_detection = False)
|
||||
#face['analysis'] = DeepFace.analyze(img, actions = ['emotion'])
|
||||
@ -279,15 +279,3 @@ with conn:
|
||||
#print(df.head())
|
||||
|
||||
update_face_count(conn, photoId, len(faces))
|
||||
exit(0)
|
||||
|
||||
#img2_path = sys.argv[2]
|
||||
#print("image 1: ", img1_path);
|
||||
#print("image 2: ", img2_path);
|
||||
#result = DeepFace.verify(img1_path = img1_path, img2_path = img2_path, #model_name = models[1])
|
||||
#print("result: ", result)
|
||||
|
||||
#face recognition
|
||||
#df = DeepFace.find(img_path = img1_path, db_path = "./db/deepface", model_name = models[1])
|
||||
|
||||
#print("df: ", df)
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user