reformating

Author: PascalReich

cropper.py

@@ -1,129 +1,56 @@
 #!/usr/bin/python
-# The contents of this file are in the public domain. See LICENSE_FOR_EXAMPLE_PROGRAMS.txt
-#
-#   This example shows how to use dlib's face recognition tool for image alignment.
-#
-# COMPILING/INSTALLING THE DLIB PYTHON INTERFACE
-#   You can install dlib using the command:
-#       pip install dlib
-#
-#   Alternatively, if you want to compile dlib yourself then go into the dlib
-#   root folder and run:
-#       python setup.py install
-#
-#   Compiling dlib should work on any operating system so long as you have
-#   CMake installed.  On Ubuntu, this can be done easily by running the
-#   command:
-#       sudo apt-get install cmake
-#
-#   Also note that this example requires Numpy which can be installed
-#   via the command:
-#       pip install numpy
-
-import sys
 
 import dlib, cv2
 import numpy as np
 
-'''if len(sys.argv) != 3:
-    print(
-        "Call this program like this:\n"
-        "   ./face_alignment.py shape_predictor_5_face_landmarks.dat ../examples/faces/bald_guys.jpg\n"
-        "You can download a trained facial shape predictor from:\n"
-        "    http://dlib.net/files/shape_predictor_5_face_landmarks.dat.bz2\n")
-    exit()'''
+predictor_path = 'models/shape_predictor_68_face_landmarks.dat'
 
-predictor_path = 'resources/shape_predictor_68_face_landmarks.dat'
-# face_file_path = 'dbs/lfw-deepfunneled/Aaron_Eckhart/Aaron_Eckhart_0001.jpg'
-
-# Load all the models we need: a detector to find the faces, a shape predictor
-# to find face landmarks so we can precisely localize the face
 detector = dlib.get_frontal_face_detector()
 sp = dlib.shape_predictor(predictor_path)
 
 
 def crop(face_file_path):
-    # Load the image using Dlib
-    # img = dlib.load_rgb_image(face_file_path)
     img = cv2.imread(face_file_path)
 
-    # Ask the detector to find the bounding boxes of each face. The 1 in the
-    # second argument indicates that we should upsample the image 1 time. This
-    # will make everything bigger and allow us to detect more faces.
     dets = detector(img, 1)
 
     num_faces = len(dets)
     if num_faces == 0:
         print("Sorry, there were no faces found in '{}'".format(face_file_path))
         return False, None
 
-    # Find the 5 face landmarks we need to do the alignment.
     faces = dlib.full_object_detections()
     for detection in dets:
         faces.append(sp(img, detection))
 
-    # window = dlib.image_window()
-
-    # Get the aligned face images
-    # Optionally:
-    # images = dlib.get_face_chips(img, faces, size=160, padding=0.25)
     images = dlib.get_face_chips(img, faces, size=160)
-    # window.set_image(image)
 
-    # TODO we can choose faces
-    # cv2.imshow("img", images[0])
-    # cv2.waitKey()
-    # cv2.imwrite(output_path, images[0])
     ret = cv2.cvtColor(images[0], cv2.COLOR_BGR2RGB)
     return img, cv2.imencode('.jpg', img)[1]
 
 
-def cropFileStorageObject(fileStorage):
-    # Load the image using Dlib
-    # img = dlib.load_rgb_image(face_file_path)
-    # read image file string data
-    filestr = fileStorage.read()
-    # convert string data to numpy array
+def cropFileStorageObject(file_storage):
+
+    filestr = file_storage.read()
+
     npimg = np.fromstring(filestr, np.uint8)
-    # convert numpy array to image
+
     img = cv2.imdecode(npimg, cv2.IMREAD_COLOR)
 
-    # Ask the detector to find the bounding boxes of each face. The 1 in the
-    # second argument indicates that we should upsample the image 1 time. This
-    # will make everything bigger and allow us to detect more faces.
     dets = detector(img, 1)
 
     num_faces = len(dets)
     if num_faces == 0:
-        print("Sorry, there were no faces found in '{}'".format(fileStorage))
+        print("Sorry, there were no faces found in '{}'".format(file_storage))
         return False, None
 
-    # Find the 5 face landmarks we need to do the alignment.
     faces = dlib.full_object_detections()
     for detection in dets:
         faces.append(sp(img, detection))
 
-    # window = dlib.image_window()
-
-    # Get the aligned face images
-    # Optionally:
-    # images = dlib.get_face_chips(img, faces, size=160, padding=0.25)
-    images = dlib.get_face_chips(img, faces, size=160)
-    # window.set_image(image)
+    # images = dlib.get_face_chips(img, faces, size=160)
 
     # TODO we can choose faces
-    # cv2.imshow("img", images[0])
-    # cv2.waitKey()
-    # cv2.imwrite(output_path, images[0])
-    ret = cv2.cvtColor(images[0], cv2.COLOR_BGR2RGB)
+    # ret = cv2.cvtColor(images[0], cv2.COLOR_BGR2RGB)
     return img, cv2.imencode('.jpg', img)[1]
 
-
-if __name__ == '__main__':
-    pass
-    # crop(face_file_path)
-
-# It is also possible to get a single chip
-# image = dlib.get_face_chip(img, faces[0])
-# window.set_image(image)
-# dlib.hit_enter_to_continue()

main.py

@@ -1,15 +1,10 @@
 from flask import Flask, render_template, request
-from werkzeug.utils import secure_filename
-import os
-import random
-import string
 import predict
 import cropper
 import time
 import base64
 
 app = Flask(__name__)
-output_path = 'temp/'
 
 
 @app.route('/')
@@ -21,20 +16,26 @@ def main():
 def upload_file():
     return render_template('upload.html')
 
+
+"""
 @app.route('/src/<folder>/<img>')
 def serve(folder, img):
     app.send_file('src/' + folder + '/' + img)
+"""
+
 
-@app.route('/uploader', methods=['GET', 'POST'])
+@app.route('/uploader', methods=['POST'])
 def file_uploaded():
     if request.method == 'POST':
         f = request.files['file']
         start = time.time()
         img, base64compat = cropper.cropFileStorageObject(f)
         labels = predict.predictFromTensor(img)
-    else: raise TypeError
-        
-    return render_template('results.html', labels=labels, base64=base64.b64encode(base64compat).decode(), time=time.time()-start)
+    else:
+        raise TypeError
+
+    return render_template('results.html', labels=labels, base64=base64.b64encode(base64compat).decode(),
+                           time=time.time() - start)
 
 
 if __name__ == '__main__':

…es/shape_predictor_68_face_landmarks.dat …ls/shape_predictor_68_face_landmarks.datresources/shape_predictor_68_face_landmarks.dat renamed to models/shape_predictor_68_face_landmarks.dat resources/shape_predictor_68_face_landmarks.dat renamed to models/shape_predictor_68_face_landmarks.dat

@@ -1,23 +1,11 @@
 import tensorflow as tf
 import numpy as np
-# import time
-# from tensorflow.keras.preprocessing.image import ImageDataGenerator
 import json, os
 
-"""
-testdatagen = ImageDataGenerator(rescale=1./255)
-
-# test_it = testdatagen.flow_from_directory('dbs/test', shuffle=False, target_size=(80, 80), class_mode='binary')
-# print(test_it.next()[0].shape)
-test_it = testdatagen.flow_from_directory('dbs/lfw-cropped', shuffle=False, target_size=(80, 80), class_mode='binary')
-
-labels = dict((v,k) for k,v in test_it.class_indices.items())
-"""
 models = {
     "id": tf.keras.models.load_model("models/id/idk-id.h5"),
     "gen": tf.keras.models.load_model("models/gen/15-altb-gen.h5"),
     "age": tf.keras.models.load_model("models/age/idk-long-age.h5")
-
 }
 
 
@@ -29,9 +17,8 @@ def get_indices(name):
 
 def get_label(name, model, img):
     pred = model.predict_on_batch(tf.expand_dims(img, 0))
-    # print(get_indices("id")[str(np.argmax(pred))])
     if name in ("id", "age"):
-        return get_indices(name)[str(np.argmax(pred))], 0  # pred[0][np.argmax(pred)].numpy()
+        return get_indices(name)[str(np.argmax(pred))], 0
     elif name == "gen":
         return get_indices("gen")[str(int(round(pred[0][0].numpy())))], pred[0][0].numpy() if not int(
             round(pred[0][0].numpy())) else 100 - pred[0][0].numpy()
@@ -60,11 +47,8 @@ def predictFromPath(img_path="dbs/test/Ariel_Sharon_0006.jpg"):
         }
     }
 
-    # print(image.shape)
-
     for mod in models.keys():
         labels = get_label(mod, models[mod], image)
-        # print(labels[0], f"{labels[1] * 100}%")
         returns[mod]["class"], returns[mod]["confidence"] = labels[0], f"{labels[1] * 100}%"
 
     return returns, img_path
@@ -89,11 +73,8 @@ def predictFromTensor(tensor):
         }
     }
 
-    # print(image.shape)
-
     for mod in models.keys():
         labels = get_label(mod, models[mod], image)
-        # print(labels[0], f"{labels[1] * 100}%")
         returns[mod]["class"], returns[mod]["confidence"] = labels[0], f"{labels[1] * 100}%"
 
     return returns