.gitignore

@@ -1,5 +1,6 @@
 data
 .idea
+.yoloface
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

face_detect.py

@@ -0,0 +1,54 @@
+import cv2
+import numpy as np
+from yoloface import face_analysis
+
+face_detector = face_analysis()
+
+
+def equalize_image(data: np.ndarray):
+  data_hsv = cv2.cvtColor(data, cv2.COLOR_RGB2HSV)
+  data_hsv[:, :, 2] = cv2.equalizeHist(data_hsv[:, :, 2])
+  return cv2.cvtColor(data_hsv, cv2.COLOR_HSV2RGB)
+
+
+def find_face_bbox_yolo(data: np.ndarray):
+  _, box, conf = face_detector.face_detection(frame_arr=data, frame_status=True, model='full')
+  if len(box) < 1:
+    return None, None
+  return box, conf
+
+
+def find_face_bbox(data: np.ndarray):
+  classifier_files = [
+    'haarcascades/haarcascade_frontalface_default.xml',
+    'haarcascades/haarcascade_frontalface_alt.xml',
+    'haarcascades/haarcascade_frontalface_alt2.xml',
+    'haarcascades/haarcascade_profileface.xml',
+    'haarcascades/haarcascade_glasses.xml',
+    'lbpcascade_animeface.xml',
+  ]
+  data_equalized = equalize_image(data)
+  data_gray = cv2.cvtColor(data_equalized, cv2.COLOR_RGB2GRAY)
+  face_coords, conf = find_face_bbox_yolo(cv2.cvtColor(data_equalized, cv2.COLOR_RGB2BGR))
+  if face_coords is not None:
+    return face_coords[0]
+
+  for classifier in classifier_files:
+    face_cascade = cv2.CascadeClassifier(classifier)
+    face_coords = face_cascade.detectMultiScale(data_gray, 1.1, 3)
+    if face_coords is not None:
+      break
+  return max(face_coords, key=lambda v: v[2]*v[3])
+
+
+def crop_face(data: np.ndarray, bounding_box) -> np.ndarray:
+  x, y, w, h = bounding_box
+  # Extending the boxes
+  factor = 0.4
+  x, y = round(x - factor * w), round(y - factor * h)
+  w, h = round(w + factor * w * 2), round(h + factor * h * 2)
+  y = max(y, 0)
+  x = max(x, 0)
+
+  face = data[y:y + h, x:x + w]
+  return face

helpers.py

@@ -0,0 +1,12 @@
+import os
+import sys
+
+
+def no_stdout(func):
+  def wrapper(*args, **kwargs):
+    old_stdout = sys.stdout
+    sys.stdout = open(os.devnull, "w")
+    ret = func(*args, **kwargs)
+    sys.stdout = old_stdout
+    return ret
+  return wrapper

load_test_data.py

@@ -5,7 +5,11 @@ import cv2 as cv
 from pathlib import Path
 
 
-def load_data(input_dir, newSize=(64,64)):
+def load_source(filename: str) -> np.ndarray:
+  return cv.imread(filename)[..., ::-1]
+
+
+def load_data(input_dir):
     image_path = Path(input_dir)
     file_names = os.listdir(image_path)
     categories_name = []
@@ -27,8 +31,7 @@ def load_data(input_dir, newSize=(64,64)):
 
     for n in file_names:
         p = image_path / n
-        img = imread(p) # zwraca ndarry postaci xSize x ySize x colorDepth
+        img = load_source(str(p)) # zwraca ndarry postaci xSize x ySize x colorDepth
-        img = cv.resize(img, newSize, interpolation=cv.INTER_AREA) # zwraca ndarray
         test_img.append(img)
         labels.append(n)
 

main.py

@@ -1,10 +1,15 @@
 import argparse
 import sys
 import cv2
+import matplotlib.pyplot as plt
 import numpy as np
 
-from comparisons import histogram_comparison, structural_similarity_index, euclidean_distance
+from metrics import histogram_comparison, structural_similarity_index, euclidean_distance, AccuracyGatherer
-from load_test_data import load_data
+
+from face_detect import find_face_bbox, crop_face
+from helpers import no_stdout
+from load_test_data import load_data, load_source
+from metrics import get_top_results
 from plots import plot_two_images, plot_results
 
 # Allows imports from the style transfer submodule
@@ -13,21 +18,10 @@ sys.path.append('DCT-Net')
 from source.cartoonize import Cartoonizer
 
 
-def load_source(filename: str) -> np.ndarray:
+anime_transfer = Cartoonizer(dataroot='DCT-Net/damo/cv_unet_person-image-cartoon_compound-models')
-  return cv2.imread(filename)[..., ::-1]
 
 
-def find_and_crop_face(data: np.ndarray, classifier_file='haarcascades/haarcascade_frontalface_default.xml') -> np.ndarray:
+def compare_with_anime_characters(source_image: np.ndarray, anime_faces_dataset: dict, verbose=False) -> list[dict]:
-  data_gray = cv2.cvtColor(data, cv2.COLOR_BGR2GRAY)
-  face_cascade = cv2.CascadeClassifier(classifier_file)
-  face = face_cascade.detectMultiScale(data_gray, 1.1, 3)
-  face = max(face, key=len)
-  x, y, w, h = face
-  face = data[y:y + h, x:x + w]
-  return face
-
-
-def compare_with_anime_characters(source: np.ndarray, anime_faces_dataset: dict, verbose=False) -> list[dict]:
   all_metrics = []
   for anime_image, label in zip(anime_faces_dataset['values'], anime_faces_dataset['labels']):
     current_result = {
@@ -37,7 +31,7 @@ def compare_with_anime_characters(source: np.ndarray, anime_faces_dataset: dict,
     # TODO: Use a different face detection method for anime images
     # anime_face = find_and_crop_face(anime_image, 'haarcascades/lbpcascade_animeface.xml')
     anime_face = anime_image
-    source_rescaled = cv2.resize(source, anime_face.shape[:2])
+    source_rescaled = cv2.resize(source_image, anime_face.shape[:2])
     if verbose:
       plot_two_images(anime_face, source_rescaled)
     current_result['metrics'] = histogram_comparison(source_rescaled, anime_face)
@@ -48,52 +42,73 @@ def compare_with_anime_characters(source: np.ndarray, anime_faces_dataset: dict,
   return all_metrics
 
 
-def get_top_results(all_metrics: list[dict], metric='correlation', count=1):
+@no_stdout
-  all_metrics.sort(reverse=True, key=lambda item: item['metrics'][metric])
-  return list(map(lambda item: {'name': item['name'], 'score': item['metrics'][metric]}, all_metrics[:count]))
-
-
 def transfer_to_anime(img: np.ndarray):
-  algo = Cartoonizer(dataroot='DCT-Net/damo/cv_unet_person-image-cartoon_compound-models')
+  model_out = anime_transfer.cartoonize(img).astype(np.uint8)
-  model_out = algo.cartoonize(img).astype(np.uint8)
   return cv2.cvtColor(model_out, cv2.COLOR_BGR2RGB)
 
 
-def validate(test_set, anime_faces_set, metric='correlation'):
+def similarity_to_anime(source_image, anime_faces_set, debug=False):
+  try:
+    source_face_bbox = find_face_bbox(source_image)
+  except ValueError:
+    return None
+  source_anime = transfer_to_anime(source_image)
+  source_face_anime = crop_face(source_anime, source_face_bbox)
+
+  if debug:
+    source_image_with_box = source_image.copy()
+    x, y, w, h = source_face_bbox
+    cv2.rectangle(source_image_with_box, (x, y), (x + w, y + h), (255, 0, 0), 2)
+    plt.figure(figsize=[12, 4])
+    plt.subplot(131)
+    plt.imshow(source_image_with_box)
+    plt.subplot(132)
+    plt.imshow(source_anime)
+    plt.subplot(133)
+    plt.imshow(source_face_anime)
+    plt.show()
+
+  return compare_with_anime_characters(source_face_anime, anime_faces_set, verbose=debug)
+
+
+def validate(test_set, anime_faces_set):
   all_entries = len(test_set['values'])
-  correct = 0
+  accuracy = AccuracyGatherer(all_entries)
   for test_image, test_label in zip(test_set['values'], test_set['labels']):
-    output = get_top_results(compare_with_anime_characters(test_image, anime_faces_set), metric)[0]['name']
+    test_results = similarity_to_anime(test_image, anime_faces_set)
-    if output == test_label:
-      correct += 1
 
-  accuracy = correct / all_entries
+    if test_results is None:
-  print(f'Accuracy using {metric}: {accuracy * 100}%')
+      print(f"cannot find face for {test_label}")
-  return accuracy
+      all_entries -= 1
+      continue
+
+    accuracy.for_results(test_results, test_label)
+
+  accuracy.count = all_entries
+  accuracy.print()
 
 
-if __name__ == '__main__':
+
+def main():
   parser = argparse.ArgumentParser()
   parser.add_argument('-v', '--validate_only')
   args = parser.parse_args()
-  anime_faces_set = load_data('data/images', (256, 256))
+  anime_faces_set = load_data('data/croped_anime_faces')
 
   if args.validate_only:
     print('Validating')
     test_set = load_data('test_set')
-    validate(test_set, anime_faces_set, 'structural-similarity')
+    validate(test_set, anime_faces_set)
-    validate(test_set, anime_faces_set, 'euclidean-distance')
-    validate(test_set, anime_faces_set, 'chi-square')
-    validate(test_set, anime_faces_set, 'correlation')
-    validate(test_set, anime_faces_set, 'intersection')
-    validate(test_set, anime_faces_set, 'bhattacharyya-distance')
     exit(0)
 
-  source = load_source('UAM-Andre.jpg')
+  source = load_source('test_set/Ayanokouji, Kiyotaka.jpg')
-  source_anime = transfer_to_anime(source)
+  results = similarity_to_anime(source, anime_faces_set)
-  source_face_anime = find_and_crop_face(source_anime)
+  method = 'correlation'
-  results = compare_with_anime_characters(source_face_anime, anime_faces_set)
-  method = 'structural-similarity'
   top_results = get_top_results(results, count=4, metric=method)
   print(top_results)
-  plot_results(source, source_anime, top_results, anime_faces_set, method)
+  plot_results(source, transfer_to_anime(source), top_results, anime_faces_set, method)
+
+
+if __name__ == '__main__':
+  main()

metrics.py

@@ -40,3 +40,42 @@ def euclidean_distance(data_a: np.ndarray, data_b: np.ndarray) -> float:
     result += (histogram_a[i] - histogram_b[i]) ** 2
     i += 1
   return result[0] ** (1 / 2)
+
+
+def get_top_results(all_metrics: list[dict], metric='correlation', count=1):
+  all_metrics.sort(reverse=True, key=lambda item: item['metrics'][metric])
+  return list(map(lambda item: {'name': item['name'], 'score': item['metrics'][metric]}, all_metrics[:count]))
+
+
+class AccuracyGatherer:
+  all_metric_names = [
+    'structural-similarity',
+    'euclidean-distance',
+    'chi-square',
+    'correlation',
+    'intersection',
+    'bhattacharyya-distance'
+  ]
+
+  def __init__(self, count, top_ks=(1, 3, 5)):
+    self.top_ks = top_ks
+    self.hits = {k: {metric: 0 for metric in AccuracyGatherer.all_metric_names} for k in top_ks}
+    self.count = count
+
+  def print(self):
+    for k in self.top_ks:
+      all_metrics = {metric: self.hits[k][metric] / self.count for metric in AccuracyGatherer.all_metric_names}
+      print(f'Top {k} matches results:')
+      [print(f'\t{key}: {value * 100}%') for key, value in all_metrics.items()]
+
+  def for_results(self, results, test_label):
+    top_results_all_metrics = {
+      k: {m: get_top_results(results, m, k) for m in AccuracyGatherer.all_metric_names} for k in self.top_ks
+    }
+    for metric_name in AccuracyGatherer.all_metric_names:
+      self.add_if_hit(top_results_all_metrics, test_label, metric_name)
+
+  def add_if_hit(self, results, test_label, metric_name):
+    for k in self.top_ks:
+      if any(map(lambda single_result: single_result['name'] == test_label, results[k][metric_name])):
+        self.hits[k][metric_name] += 1

requirements.txt

@@ -8,4 +8,6 @@ lxml==4.9.2
 opencv-python==4.7.0.68
 torch==1.13.1
 matplotlib==3.6.3
-scikit-image==0.19.3
+scikit-image==0.19.3
+yoloface==0.0.4
+ipython==8.9.0