Merge pull request 'comparison' (#3) from comparison into main

Reviewed-on: #3

.gitignore

@@ -1,4 +1,5 @@
 data
+.idea
 
 # Byte-compiled / optimized / DLL files
 __pycache__/

comparisons.py

@@ -0,0 +1,42 @@
+import cv2
+import numpy as np
+from skimage.metrics import structural_similarity
+
+def histogram_comparison(data_a: np.ndarray, data_b: np.ndarray) -> dict:
+  hsv_a = cv2.cvtColor(data_a, cv2.COLOR_BGR2HSV)
+  hsv_b = cv2.cvtColor(data_b, cv2.COLOR_BGR2HSV)
+
+  histSize = [50, 60]
+  hue_ranges = [0, 180]
+  sat_ranges = [0, 256]
+  channels = [0, 1]
+  ranges = hue_ranges + sat_ranges
+
+  hist_a = cv2.calcHist([hsv_a], channels, None, histSize, ranges, accumulate=False)
+  cv2.normalize(hist_a, hist_a, alpha=0, beta=1, norm_type=cv2.NORM_MINMAX)
+  hist_b = cv2.calcHist([hsv_b], channels, None, histSize, ranges, accumulate=False)
+  cv2.normalize(hist_b, hist_b, alpha=0, beta=1, norm_type=cv2.NORM_MINMAX)
+
+  return {
+    'correlation': cv2.compareHist(hist_a, hist_b, 0),
+    'chi-square': cv2.compareHist(hist_a, hist_b, 1),
+    'intersection': cv2.compareHist(hist_a, hist_b, 2),
+    'bhattacharyya-distance': cv2.compareHist(hist_a, hist_b, 3),
+  }
+
+
+def structural_similarity_index(data_a: np.ndarray, data_b: np.ndarray) -> float:
+  return structural_similarity(cv2.cvtColor(data_a, cv2.COLOR_BGR2GRAY), cv2.cvtColor(data_b, cv2.COLOR_BGR2GRAY))
+
+
+def euclidean_distance(data_a: np.ndarray, data_b: np.ndarray) -> float:
+  gray_a = cv2.cvtColor(data_a, cv2.COLOR_BGR2GRAY)
+  histogram_a = cv2.calcHist([gray_a], [0], None, [256], [0, 256])
+
+  gray_b = cv2.cvtColor(data_b, cv2.COLOR_BGR2GRAY)
+  histogram_b = cv2.calcHist([gray_b], [0],  None, [256], [0, 256])
+  result, i = [0.], 0
+  while i < len(histogram_a) and i < len(histogram_b):
+    result += (histogram_a[i] - histogram_b[i]) ** 2
+    i += 1
+  return result[0] ** (1 / 2)

main.py

@@ -1,40 +1,60 @@
-# Allows imports from the style transfer submodule
 import sys
-sys.path.append('DCT-Net')
-
 import cv2
-import os
 import numpy as np
+import matplotlib.pyplot as plt
+
+from comparisons import histogram_comparison, structural_similarity_index, euclidean_distance
+
+# Allows imports from the style transfer submodule
+sys.path.append('DCT-Net')
 
 from source.cartoonize import Cartoonizer
 
 
 def load_source(filename: str) -> np.ndarray:
-  return cv2.imread(filename)[...,::-1]
+  return cv2.imread(filename)[..., ::-1]
 
 
-def find_and_crop_face(data: np.ndarray) -> np.ndarray:
+def find_and_crop_face(data: np.ndarray, classifier_file='haarcascades/haarcascade_frontalface_default.xml') -> np.ndarray:
   data_gray = cv2.cvtColor(data, cv2.COLOR_BGR2GRAY)
-  face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_default.xml')
-  face = face_cascade.detectMultiScale(data_gray, 1.3, 4)
+  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 plot_two_images(a: np.ndarray, b: np.ndarray):
+  plt.figure(figsize=[10, 10])
+  plt.subplot(121)
+  plt.imshow(a)
+  plt.title("A")
+  plt.subplot(122)
+  plt.imshow(b)
+  plt.title("B")
+  plt.show()
+
+
 def compare_with_anime_characters(data: np.ndarray) -> int:
-  # TODO
-  return 1
+  # Example will be one face from anime dataset
+  example = load_source('data/images/Aisaka, Taiga.jpg')
+  # TODO: Use a different face detection method for anime images
+  example_face = find_and_crop_face(example, 'haarcascades/lbpcascade_animeface.xml')
+  data_rescaled = cv2.resize(data, example_face.shape[:2])
+  plot_two_images(example_face, data_rescaled)
+  print(histogram_comparison(data_rescaled, example_face))
+  print(f'structural-similarity: {structural_similarity_index(data_rescaled, example_face)}')
+  print(f'euclidean-distance: {euclidean_distance(data_rescaled, example_face)}')
 
 
 def transfer_to_anime(img: np.ndarray):
-  algo = Cartoonizer(dataroot='damo/cv_unet_person-image-cartoon_compound-models')
-  return algo.cartoonize(img)
+  algo = Cartoonizer(dataroot='DCT-Net/damo/cv_unet_person-image-cartoon_compound-models')
+  return algo.cartoonize(img).astype(np.uint8)
 
 
 if __name__ == '__main__':
-  source = load_source('input.png')
-  source_face = find_and_crop_face(source)
-  source_face_anime = transfer_to_anime(source)
+  source = load_source('UAM-Andre.jpg')
+  source_anime = transfer_to_anime(source)
+  source_face_anime = find_and_crop_face(source_anime)
   print(compare_with_anime_characters(source_face_anime))

requirements.txt

@@ -5,4 +5,7 @@ modelscope==1.1.3
 requests==2.28.2
 beautifulsoup4==4.11.1
 lxml==4.9.2
-opencv-python==4.7.0.68
+opencv-python==4.7.0.68
+torch==1.13.1
+matplotlib==3.6.3
+scikit-image==0.19.3