def upload():
target = os.path.join(APP_ROOT, "images/")
print(target)
if not os.path.isdir(target):
os.mkdir(target)
i = 0
for file in request.files.getlist("file"):
print(file)
filename = file.filename
destination = "/".join([target, "a" + str(i) + ".png"])
i += 1
print(destination)
file.save(destination)
save_face1("D:\T\SpitHackathon\src\images/a0.png")
img_1 = readImage("D:\T\SpitHackathon\src\images/a.jpeg")
save_face2("D:\T\SpitHackathon\src\data/frame0.jpg")
img_2 = readImage("D:\T\SpitHackathon\src\images/b.jpeg")
output = verify_images(img_1, img_2)
if (output):
return render_template("complete.html")
else:
return render_template("notComplete.html")
def getConvImage(name, layer=2, index=0):
image = model_images.readImage(name, cv_2=True)
image = model_images.getFaces(image)[0]
images = getConvList(image, layer=8)
img = images[index]
img = Image.fromarray(img)
return img
def extract_faces(source=None, destination=None):
if (source == None): source = os.path.join(os.getcwd(), 'lfw_data')
if (destination == None):
destination = os.path.join(os.getcwd(), 'lfw_data_faces')
for folder in os.listdir(source):
folder_path = os.path.join(source, folder)
os.mkdir(os.path.join(destination, folder))
for img in os.listdir(folder_path):
img_path = os.path.join(folder_path, img)
print(img_path)
faces = getFaces(readImage(img_path, cv_2=True))
if len(faces) > 0:
face = faces[0]
face.save(os.path.join(destination, folder, img))
def generate_np_matrix(source=None):
print('Creating Numpy Matrix...')
if (source == None): source = os.path.join(os.getcwd(), 'lfw')
img_arr = []
for folder in os.listdir(source):
folder_path = os.path.join(source, folder)
for img in os.listdir(folder_path):
img_path = os.path.join(folder_path, img)
image = readImage(img_path)
ima = np.array(image)
img_arr.append(ima)
np_arr = np.array(img_arr)
print('Numpy Matrix Created.')
return np_arr
from model_images import getFaces, readImage
import numpy as np
import cv2
from PIL import Image
import time
import pickle
import os
import pandas as pd
loaded_model = pickle.load(open('model.sav', 'rb'))
while True:
tic = time.time()
IMG = readImage('download.jpg', cv_2=True)
faces = getFaces(IMG)
if len(faces) != 0:
face = faces[0]
vec = getFeatureVector(face)
cluster = loaded_model.predict([vec])
path = os.path.join(os.getcwd(), 'Clusters')
cl_path = os.path.join(path, 'Cluster_' + str(cluster[0]),
'cluster_' + str(cluster[0]) + '_data.csv')
df = pd.read_csv(cl_path)
vectors = df.to_numpy()
for r_idx, og_vec in enumerate(vectors):
import os
import numpy as np
from PIL import Image
from model_images import getFaces, highlightFaces, readImage, cv2_to_pil
path = os.path.join(os.getcwd(), 'IMDB_Data')
new_path = os.path.join(os.getcwd(), 'Data')
file_no = 0
for file in os.listdir(path):
print('Processing file: ', file_no, '\n')
img_dir = os.path.join(path, file)
list_images = os.listdir(img_dir)
new_file = os.path.join(new_path, 'file_' + str(file_no))
os.mkdir(new_file)
img_no = 0
for image in list_images:
img = readImage(os.path.join(img_dir, image), cv_2=True)
images = getFaces(img)
if len(images) != 0:
img = images[0]
name = str(img_no) + '.jpg'
save_path = os.path.join(new_file, name)
img.save(save_path)
img_no += 1
file_no += 1
arr = np.empty((0,2622), float)
def generate_dataframe(data):
df = pd.DataFrame(data, columns = feature_col_idx)
return df
def appendImgToNPA(image, arr, name):
print('Processing image ', name)
vec = getFeatureVector(image)
arr = np.append(arr, [vec], axis = 0)
return arr
'''
for filename in os.listdir(path):
img_path = os.path.join(path, filename)
image = readImage(img_path)
img = np.asarray(image)
image_np = np.empty(img.shape, float)
image_np = np.append(image_np, img, axis = 0)
image_np = np.append(image_np, img, axis = 0)
image = Image.fromarray(image_np)
arr = appendImgToNPA(image, arr, filename)
'''
for filename in img_list:
path = os.path.join(os.getcwd(), 'Yale', filename)
img_path = os.path.join(path)
image = readImage(img_path)
arr = appendImgToNPA(image, arr, path)
df = generate_dataframe(arr)
df.to_csv(r'Yale_DataFrame.csv', index = False)