File size: 5,137 Bytes
be751d2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
# Modified from https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/flux2/image_processor.py
# Copyright 2025 The Black Forest Labs Team and The HuggingFace 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.

import math
from typing import Tuple

import PIL.Image

from diffusers.configuration_utils import register_to_config
from diffusers.image_processor import VaeImageProcessor


class Flux2ImageProcessor(VaeImageProcessor):
    r"""
    Image processor to preprocess the reference (character) image for the Flux2 model.

    Args:
        do_resize (`bool`, *optional*, defaults to `True`):
            Whether to downscale the image's (height, width) dimensions to multiples of `vae_scale_factor`. Can accept
            `height` and `width` arguments from [`image_processor.VaeImageProcessor.preprocess`] method.
        vae_scale_factor (`int`, *optional*, defaults to `16`):
            VAE (spatial) scale factor. If `do_resize` is `True`, the image is automatically resized to multiples of
            this factor.
        vae_latent_channels (`int`, *optional*, defaults to `32`):
            VAE latent channels.
        do_normalize (`bool`, *optional*, defaults to `True`):
            Whether to normalize the image to [-1,1].
        do_convert_rgb (`bool`, *optional*, defaults to be `True`):
            Whether to convert the images to RGB format.
    """

    @register_to_config
    def __init__(
        self,
        do_resize: bool = True,
        vae_scale_factor: int = 16,
        vae_latent_channels: int = 32,
        do_normalize: bool = True,
        do_convert_rgb: bool = True,
    ):
        super().__init__(
            do_resize=do_resize,
            vae_scale_factor=vae_scale_factor,
            vae_latent_channels=vae_latent_channels,
            do_normalize=do_normalize,
            do_convert_rgb=do_convert_rgb,
        )

    @staticmethod
    def check_image_input(
        image: PIL.Image.Image, max_aspect_ratio: int = 8, min_side_length: int = 64, max_area: int = 1024 * 1024
    ) -> PIL.Image.Image:
        """
        Check if image meets minimum size and aspect ratio requirements.

        Args:
            image: PIL Image to validate
            max_aspect_ratio: Maximum allowed aspect ratio (width/height or height/width)
            min_side_length: Minimum pixels required for width and height
            max_area: Maximum allowed area in pixels²

        Returns:
            The input image if valid

        Raises:
            ValueError: If image is too small or aspect ratio is too extreme
        """
        if not isinstance(image, PIL.Image.Image):
            raise ValueError(f"Image must be a PIL.Image.Image, got {type(image)}")

        width, height = image.size

        # Check minimum dimensions
        if width < min_side_length or height < min_side_length:
            raise ValueError(
                f"Image too small: {width}×{height}. Both dimensions must be at least {min_side_length}px"
            )

        # Check aspect ratio
        aspect_ratio = max(width / height, height / width)
        if aspect_ratio > max_aspect_ratio:
            raise ValueError(
                f"Aspect ratio too extreme: {width}×{height} (ratio: {aspect_ratio:.1f}:1). "
                f"Maximum allowed ratio is {max_aspect_ratio}:1"
            )

        return image

    @staticmethod
    def _resize_to_target_area(image: PIL.Image.Image, target_area: int = 1024 * 1024) -> Tuple[int, int]:
        image_width, image_height = image.size

        scale = math.sqrt(target_area / (image_width * image_height))
        width = int(image_width * scale)
        height = int(image_height * scale)

        return image.resize((width, height), PIL.Image.Resampling.LANCZOS)

    def _resize_and_crop(
        self,
        image: PIL.Image.Image,
        width: int,
        height: int,
    ) -> PIL.Image.Image:
        r"""
        center crop the image to the specified width and height.

        Args:
            image (`PIL.Image.Image`):
                The image to resize and crop.
            width (`int`):
                The width to resize the image to.
            height (`int`):
                The height to resize the image to.

        Returns:
            `PIL.Image.Image`:
                The resized and cropped image.
        """
        image_width, image_height = image.size

        left = (image_width - width) // 2
        top = (image_height - height) // 2
        right = left + width
        bottom = top + height

        return image.crop((left, top, right, bottom))