utils.py

""" Utility classes and functions related to FRES (ACL 2025).
Copyright (c) 2025 Robert Bosch GmbH
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published
by the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
butWITHOUT ANYWARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
"""

import torchvision.transforms as T
from torchvision.transforms.functional import InterpolationMode
from PIL import Image
import torch


def load_image(image_file, input_size=448, max_num=2):
    image = Image.open(image_file).convert('RGB')
    images = dynamic_preprocess(
        image, image_size=input_size, use_thumbnail=True, max_num=max_num)
    transform = build_transform(input_size=input_size)
    pixel_values = [transform(image) for image in images]
    pixel_values = torch.stack(pixel_values)
    return pixel_values


def build_transform(input_size):
    IMAGENET_MEAN = (0.485, 0.456, 0.406)
    IMAGENET_STD = (0.229, 0.224, 0.225)
    MEAN, STD = IMAGENET_MEAN, IMAGENET_STD
    transform = T.Compose([
        T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
        T.Resize((input_size, input_size),
                 interpolation=InterpolationMode.BICUBIC),
        T.ToTensor(),
        T.Normalize(mean=MEAN, std=STD)
    ])
    return transform


def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
    best_ratio_diff = float('inf')
    best_ratio = (1, 1)
    area = width * height
    for ratio in target_ratios:
        target_aspect_ratio = ratio[0] / ratio[1]
        ratio_diff = abs(aspect_ratio - target_aspect_ratio)
        if ratio_diff < best_ratio_diff:
            best_ratio_diff = ratio_diff
            best_ratio = ratio
        elif ratio_diff == best_ratio_diff:
            if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
                best_ratio = ratio
    return best_ratio


def dynamic_preprocess(image, min_num=1, max_num=12, image_size=448, use_thumbnail=False):
    orig_width, orig_height = image.size
    aspect_ratio = orig_width / orig_height

    # calculate the existing image aspect ratio
    target_ratios = set(
        (i, j) for n in range(min_num, max_num + 1) for i in range(1, n + 1) for j in range(1, n + 1) if
        i * j <= max_num and i * j >= min_num)
    target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])

    # find the closest aspect ratio to the target
    target_aspect_ratio = find_closest_aspect_ratio(
        aspect_ratio, target_ratios, orig_width, orig_height, image_size)

    # calculate the target width and height
    target_width = image_size * target_aspect_ratio[0]
    target_height = image_size * target_aspect_ratio[1]
    blocks = target_aspect_ratio[0] * target_aspect_ratio[1]

    # resize the image
    resized_img = image.resize((target_width, target_height))
    processed_images = []
    for i in range(blocks):
        box = (
            (i % (target_width // image_size)) * image_size,
            (i // (target_width // image_size)) * image_size,
            ((i % (target_width // image_size)) + 1) * image_size,
            ((i // (target_width // image_size)) + 1) * image_size
        )
        # split the image
        split_img = resized_img.crop(box)
        processed_images.append(split_img)
    assert len(processed_images) == blocks
    if use_thumbnail and len(processed_images) != 1:
        thumbnail_img = image.resize((image_size, image_size))
        processed_images.append(thumbnail_img)
    return processed_images


def reformat_table(table, caption, style=0):
    # reformat table list into str
    output = ""
    if caption:
        output += f"Table title: {caption}\nTable:"
    if not isinstance(table, dict):
        # tat format
        header = table[0]
        rows = table[1:]
    else:
        header = table['header']
        rows = table['rows']
    if style == 0:
        output = ' | '.join(header)
        output += '\n'
        for row in rows:
            output += ' | '.join(row)
            output += '\n'
    elif style == 1:
        output = str(header)
        output += ','
        for row in rows:
            output += str(row)
            output += ','
        output = output[:-1]
    elif style == 2:
        output = ",".join(header)
        for i, row in enumerate(rows):
            output += f",[ROW {i+1}]," + ",".join(row)
    return output