model/multimodal_encoder/siglip/image_processing_flashsloth.py

# coding=utf-8
# Copyright 2024 Zhenwei Shao and MILVLG team.
# Licensed under the Apache License, Version 2.0.

# Adopted from https://github.com/huggingface/transformers/tree/main/src/transformers/models/siglip. 
# Below is the original copyright:
# Copyright 2022 The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Image processor class for FlashSloth."""

from typing import Dict, List, Optional, Union

import numpy as np

from transformers.image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from transformers.image_transforms import (
    center_crop,
    convert_to_rgb,
    get_resize_output_image_size,
    normalize,
    rescale,
    resize,
    to_channel_dimension_format,
)
from transformers.image_utils import (
    ChannelDimension,
    ImageInput,
    make_list_of_images,
    to_numpy_array,
    valid_images,
)
from transformers.utils import TensorType

import PIL
from PIL.Image import Resampling as PILImageResampling


class ImpImageProcessor(BaseImageProcessor):
    r"""
    Constructs a CLIP image processor.

    Args:
        do_resize (`bool`, *optional*, defaults to `True`):
            Whether to resize the image's (height, width) dimensions to the specified `size`. Can be overridden by
            `do_resize` in the `preprocess` method.
        size (`Dict[str, int]` *optional*, defaults to `{"shortest_edge": 224}`):
            Size of the image after resizing. The shortest edge of the image is resized to size["shortest_edge"], with
            the longest edge resized to keep the input aspect ratio. Can be overridden by `size` in the `preprocess`
            method.
        resample (`PILImageResampling`, *optional*, defaults to `PILImageResampling.BICUBIC`):
            Resampling filter to use if resizing the image. Can be overridden by `resample` in the `preprocess` method.
        do_center_crop (`bool`, *optional*, defaults to `True`):
            Whether to center crop the image to the specified `crop_size`. Can be overridden by `do_center_crop` in the
            `preprocess` method.
        crop_size (`Dict[str, int]` *optional*, defaults to 224):
            Size of the output image after applying `center_crop`. Can be overridden by `crop_size` in the `preprocess`
            method.
        do_rescale (`bool`, *optional*, defaults to `True`):
            Whether to rescale the image by the specified scale `rescale_factor`. Can be overridden by `do_rescale` in
            the `preprocess` method.
        rescale_factor (`int` or `float`, *optional*, defaults to `1/255`):
            Scale factor to use if rescaling the image. Can be overridden by `rescale_factor` in the `preprocess`
            method.
        do_normalize:
            Whether to normalize the image. Can be overridden by `do_normalize` in the `preprocess` method.
        image_mean (`float` or `List[float]`, *optional*, defaults to `[0.48145466, 0.4578275, 0.40821073]`):
            Mean to use if normalizing the image. This is a float or list of floats the length of the number of
            channels in the image. Can be overridden by the `image_mean` parameter in the `preprocess` method.
        image_std (`float` or `List[float]`, *optional*, defaults to `[0.26862954, 0.26130258, 0.27577711]`):
            Image standard deviation.
        do_convert_rgb (`bool`, *optional*, defaults to `True`):
            Standard deviation to use if normalizing the image. This is a float or list of floats the length of the
            number of channels in the image. Can be overridden by the `image_std` parameter in the `preprocess` method.
    """

    model_input_names = ["pixel_values"]

    def __init__(
        self,
        do_resize: bool = True,
        size: Dict[str, int] = None,
        resample: PILImageResampling = PILImageResampling.BICUBIC,
        do_center_crop: bool = False,
        crop_size: Dict[str, int] = None,
        do_rescale: bool = True,
        rescale_factor: Union[int, float] = 1 / 255,
        do_normalize: bool = True,
        image_mean: Optional[Union[float, List[float]]] = None,
        image_std: Optional[Union[float, List[float]]] = None,
        do_convert_rgb: bool = True,
        if_squash: bool = True,
        **kwargs,
    ) -> None:
        super().__init__(**kwargs)
        size = size if size is not None else {"shortest_edge": 384}
        size = get_size_dict(size, default_to_square=False)
        crop_size = crop_size if crop_size is not None else {"height": 384, "width": 384}
        crop_size = get_size_dict(crop_size, default_to_square=True, param_name="crop_size")

        self.do_resize = do_resize
        self.size = size
        self.resample = resample
        self.do_center_crop = do_center_crop
        self.crop_size = crop_size
        self.do_rescale = do_rescale
        self.rescale_factor = rescale_factor
        self.do_normalize = do_normalize
        self.image_mean = image_mean if image_mean is not None else (0.5, 0.5, 0.5)
        self.image_std = image_std if image_std is not None else (0.5, 0.5, 0.5)
        self.do_convert_rgb = do_convert_rgb
        self.if_squash = if_squash

    def resize(
        self,
        image: np.ndarray,
        size: Dict[str, int],
        resample: PILImageResampling = PILImageResampling.BICUBIC,
        data_format: Optional[Union[str, ChannelDimension]] = None,
        **kwargs,
    ) -> np.ndarray:
        """
        Resize an image. The shortest edge of the image is resized to size["shortest_edge"], with the longest edge
        resized to keep the input aspect ratio, when `if_squash` is `False`.
        Otherwise, squash the image into a square of size `size["shortest_edge"]`.
        """
        size = get_size_dict(size, default_to_square=False)
        if "shortest_edge" not in size:
            raise ValueError(f"The `size` parameter must contain the key `shortest_edge`. Got {size.keys()}")
        output_size = get_resize_output_image_size(image, size=size["shortest_edge"], default_to_square=self.if_squash)
        return resize(image, size=output_size, resample=resample, data_format=data_format, **kwargs)

    def center_crop(
        self,
        image: np.ndarray,
        size: Dict[str, int],
        data_format: Optional[Union[str, ChannelDimension]] = None,
        **kwargs,
    ) -> np.ndarray:
        """
        Center crop an image. If the image is too small to be cropped to the size given, it will be padded (so the
        returned result will always be of size `size`).
        """
        size = get_size_dict(size)
        if "height" not in size or "width" not in size:
            raise ValueError(f"The `size` parameter must contain the keys (height, width). Got {size.keys()}")
        return center_crop(image, size=(size["height"], size["width"]), data_format=data_format, **kwargs)

    def rescale(
        self,
        image: np.ndarray,
        scale: Union[int, float],
        data_format: Optional[Union[str, ChannelDimension]] = None,
        **kwargs,
    ):
        """
        Rescale an image by a scale factor. image = image * scale.
        """
        return rescale(image, scale=scale, data_format=data_format, **kwargs)

    def normalize(
        self,
        image: np.ndarray,
        mean: Union[float, List[float]],
        std: Union[float, List[float]],
        data_format: Optional[Union[str, ChannelDimension]] = None,
        **kwargs,
    ) -> np.ndarray:
        """
        Normalize an image. image = (image - image_mean) / image_std.
        """
        return normalize(image, mean=mean, std=std, data_format=data_format, **kwargs)

    def preprocess(
        self,
        images: ImageInput,
        do_resize: bool = None,
        size: Dict[str, int] = None,
        resample: PILImageResampling = None,
        do_center_crop: bool = None,
        crop_size: int = None,
        do_rescale: bool = None,
        rescale_factor: float = None,
        do_normalize: bool = None,
        image_mean: Optional[Union[float, List[float]]] = None,
        image_std: Optional[Union[float, List[float]]] = None,
        do_convert_rgb: bool = None,
        return_tensors: Optional[Union[str, TensorType]] = None,
        data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
        **kwargs,
    ) -> PIL.Image.Image:
        """
        Preprocess an image or batch of images.

        Args:
            images (`ImageInput`):
                Image to preprocess.
            do_resize (`bool`, *optional*, defaults to `self.do_resize`):
                Whether to resize the image.
            size (`Dict[str, int]`, *optional*, defaults to `self.size`):
                Size of the image after resizing. Shortest edge of the image is resized to size["shortest_edge"], with
                the longest edge resized to keep the input aspect ratio.
            resample (`int`, *optional*, defaults to `self.resample`):
                Resampling filter to use if resizing the image. This can be one of the enum `PILImageResampling`. Only
                has an effect if `do_resize` is set to `True`.
            do_center_crop (`bool`, *optional*, defaults to `self.do_center_crop`):
                Whether to center crop the image.
            crop_size (`Dict[str, int]`, *optional*, defaults to `self.crop_size`):
                Size of the center crop. Only has an effect if `do_center_crop` is set to `True`.
            do_rescale (`bool`, *optional*, defaults to `self.do_rescale`):
                Whether to rescale the image.
            rescale_factor (`float`, *optional*, defaults to `self.rescale_factor`):
                Rescale factor to rescale the image by if `do_rescale` is set to `True`.
            do_normalize (`bool`, *optional*, defaults to `self.do_normalize`):
                Whether to normalize the image.
            image_mean (`float` or `List[float]`, *optional*, defaults to `self.image_mean`):
                Image mean to use for normalization. Only has an effect if `do_normalize` is set to `True`.
            image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`):
                Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to
                `True`.
            do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`):
                Whether to convert the image to RGB.
            return_tensors (`str` or `TensorType`, *optional*):
                The type of tensors to return. Can be one of:
                - Unset: Return a list of `np.ndarray`.
                - `TensorType.TENSORFLOW` or `'tf'`: Return a batch of type `tf.Tensor`.
                - `TensorType.PYTORCH` or `'pt'`: Return a batch of type `torch.Tensor`.
                - `TensorType.NUMPY` or `'np'`: Return a batch of type `np.ndarray`.
                - `TensorType.JAX` or `'jax'`: Return a batch of type `jax.numpy.ndarray`.
            data_format (`ChannelDimension` or `str`, *optional*, defaults to `ChannelDimension.FIRST`):
                The channel dimension format for the output image. Can be one of:
                - `ChannelDimension.FIRST`: image in (num_channels, height, width) format.
                - `ChannelDimension.LAST`: image in (height, width, num_channels) format.
                - Unset: defaults to the channel dimension format of the input image.
        """
        do_resize = do_resize if do_resize is not None else self.do_resize
        size = size if size is not None else self.size
        size = get_size_dict(size, param_name="size", default_to_square=False)
        resample = resample if resample is not None else self.resample
        do_center_crop = do_center_crop if do_center_crop is not None else self.do_center_crop
        crop_size = crop_size if crop_size is not None else self.crop_size
        crop_size = get_size_dict(crop_size, param_name="crop_size", default_to_square=True)
        do_rescale = do_rescale if do_rescale is not None else self.do_rescale
        rescale_factor = rescale_factor if rescale_factor is not None else self.rescale_factor
        do_normalize = do_normalize if do_normalize is not None else self.do_normalize
        image_mean = image_mean if image_mean is not None else self.image_mean
        image_std = image_std if image_std is not None else self.image_std
        do_convert_rgb = do_convert_rgb if do_convert_rgb is not None else self.do_convert_rgb

        images = make_list_of_images(images)

        if not valid_images(images):
            raise ValueError(
                "Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, "
                "torch.Tensor, tf.Tensor or jax.ndarray."
            )

        if do_resize and size is None:
            raise ValueError("Size must be specified if do_resize is True.")

        if do_center_crop and crop_size is None:
            raise ValueError("Crop size must be specified if do_center_crop is True.")

        if do_rescale and rescale_factor is None:
            raise ValueError("Rescale factor must be specified if do_rescale is True.")

        if do_normalize and (image_mean is None or image_std is None):
            raise ValueError("Image mean and std must be specified if do_normalize is True.")

        # PIL RGBA images are converted to RGB
        if do_convert_rgb:
            images = [convert_to_rgb(image) for image in images]

        # All transformations expect numpy arrays.
        images = [to_numpy_array(image) for image in images]

        if do_resize:
            images = [self.resize(image=image, size=size, resample=resample) for image in images]

        if do_center_crop:
            images = [self.center_crop(image=image, size=crop_size) for image in images]

        if do_rescale:
            images = [self.rescale(image=image, scale=rescale_factor) for image in images]

        if do_normalize:
            images = [self.normalize(image=image, mean=image_mean, std=image_std) for image in images]

        images = [to_channel_dimension_format(image, data_format) for image in images]

        data = {"pixel_values": images}
        return BatchFeature(data=data, tensor_type=return_tensors)


# from transformers.image_processing_utils import BatchFeature
# from transformers.image_transforms import (
#     convert_to_rgb,
#     normalize,
#     rescale,
#     resize,
#     to_channel_dimension_format,
# )
# from transformers.image_utils import (
#     ChannelDimension,
#     PILImageResampling,
#     to_numpy_array,
# )

from PIL import Image

from functools import partial, reduce

def simple_image_processor(
        images, 
        image_mean=(0.5, 0.5, 0.5), 
        image_std=(0.5, 0.5, 0.5), 
        size=(384, 384), 
        resample=PILImageResampling.BICUBIC, 
        rescale_factor=1 / 255, 
        data_format=ChannelDimension.FIRST,
        return_tensors="pt"
    ):

    if isinstance(images, Image.Image):
        images = [images]
    else:
        assert isinstance(images, list)
    
    transforms = [
        convert_to_rgb,
        to_numpy_array,
        partial(resize, size=size, resample=resample, data_format=data_format),
        partial(rescale, scale=rescale_factor, data_format=data_format),
        partial(normalize, mean=image_mean, std=image_std, data_format=data_format),
        partial(to_channel_dimension_format, channel_dim=data_format, input_channel_dim=data_format),
    ]

    images = reduce(lambda x, f: [*map(f, x)], transforms, images)
    data = {"pixel_values": images}
    
    return BatchFeature(data=data, tensor_type=return_tensors)