def preprocess_img(img, target_image_size):
s = min(img.size)
r = target_image_size / s
s = (round(r * img.size[1]), round(r * img.size[0]))
img = TF.resize(img, s, interpolation=PIL.Image.LANCZOS)
# img = TF.center_crop(img, output_size=2 * [target_image_size])
img = torch.unsqueeze(T.ToTensor()(img), 0)
return map_pixels(img)
def preprocessing(img):
s = min(img.size)
if s < target_size:
raise ValueError(f'min dim for image')
r = target_size / s
s = round(r * img.size[1]),round(r * img.size[0])
img = TF.resize(img,s,interpolation=PIL.Image.LANCZOS)
img = TF.center_crop(img,output_size=2*[target_size])
img = torch.unsqueeze(T.ToTensor()(img),0)
return map_pixels(img)
def preprocess(img):
min_img_dim = min(img.size)
if min_img_dim < target_img_size:
raise ValueError(f'min dim for img {min_img_dim} < {target_img_size}')
img_ratio = target_img_size / min_img_dim
min_img_dim = (round(img_ratio * img.size[1]),
round(img_ratio * img.size[0]))
img = TF.resize(img, min_img_dim, interpolation=PIL.Image.LANCZOS)
img = TF.center_crop(img, output_size=2 * [target_img_size])
img = torch.unsqueeze(T.ToTensor()(img), 0)
return map_pixels(img)