import ast import math from PIL import Image def get_anyres_image_grid_shape(image_size, grid_pinpoints, patch_size): """ Calculate the shape of the image patch grid after the preprocessing for images of any resolution. Args: image_size (tuple): The size of the input image in the format (width, height). grid_pinpoints (str): A string representation of a list of possible resolutions. patch_size (int): The size of each image patch. Returns: tuple: The shape of the image patch grid in the format (width, height). """ if type(grid_pinpoints) is list: possible_resolutions = grid_pinpoints else: possible_resolutions = ast.literal_eval(grid_pinpoints) width, height = select_best_resolution(image_size, possible_resolutions) return width // patch_size, height // patch_size def select_best_resolution(original_size, possible_resolutions): """ Selects the best resolution from a list of possible resolutions based on the original size. Args: original_size (tuple): The original size of the image in the format (width, height). possible_resolutions (list): A list of possible resolutions in the format [(width1, height1), (width2, height2), ...]. Returns: tuple: The best fit resolution in the format (width, height). """ original_width, original_height = original_size best_fit = None max_effective_resolution = 0 min_wasted_resolution = float('inf') for width, height in possible_resolutions: scale = min(width / original_width, height / original_height) downscaled_width, downscaled_height = int(original_width * scale), int(original_height * scale) effective_resolution = min(downscaled_width * downscaled_height, original_width * original_height) wasted_resolution = (width * height) - effective_resolution if effective_resolution > max_effective_resolution or (effective_resolution == max_effective_resolution and wasted_resolution < min_wasted_resolution): max_effective_resolution = effective_resolution min_wasted_resolution = wasted_resolution best_fit = (width, height) return best_fit ## added by llava-1.6 def divide_to_patches(image, patch_size): """ Divides an image into patches of a specified size. Args: image (PIL.Image.Image): The input image. patch_size (int): The size of each patch. Returns: list: A list of PIL.Image.Image objects representing the patches. """ patches = [] width, height = image.size for i in range(0, height, patch_size): for j in range(0, width, patch_size): box = (j, i, j + patch_size, i + patch_size) patch = image.crop(box) patches.append(patch) return patches ## added by llava-1.6 def resize_and_pad_image(image, target_resolution): """ Resize and pad an image to a target resolution while maintaining aspect ratio. Args: image (PIL.Image.Image): The input image. target_resolution (tuple): The target resolution (width, height) of the image. Returns: PIL.Image.Image: The resized and padded image. """ original_width, original_height = image.size target_width, target_height = target_resolution scale_w = target_width / original_width scale_h = target_height / original_height if scale_w < scale_h: new_width = target_width new_height = min(math.ceil(original_height * scale_w), target_height) else: new_height = target_height new_width = min(math.ceil(original_width * scale_h), target_width) # Resize the image resized_image = image.resize((new_width, new_height)) new_image = Image.new('RGB', (target_width, target_height), (0, 0, 0)) paste_x = (target_width - new_width) // 2 paste_y = (target_height - new_height) // 2 new_image.paste(resized_image, (paste_x, paste_y)) return new_image def get_value_from_kwargs(kwargs, name): if name in kwargs: return kwargs.pop(name) else: return None