File size: 2,827 Bytes
32e2fdf
 
 
4dbde44
32e2fdf
 
 
 
 
 
661eb08
32e2fdf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bd76182
32e2fdf
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
#!/usr/bin/env python3
import numpy as np
import torch
import cv2

from transformers.tools.base import Tool, get_default_device
from transformers.utils import (
    is_accelerate_available,
    is_vision_available,
)
from diffusers import StableDiffusionControlNetPipeline, ControlNetModel, UniPCMultistepScheduler


if is_vision_available():
    from PIL import Image



IMAGE_TRANSFORMATION_DESCRIPTION = (
    "This is a tool that transforms an image with ControlNet according to a prompt. It takes two inputs: `image`, which should be "
    "the image to transform, and `prompt`, which should be the prompt to use to change it. It returns the "
    "modified image."
)


class ControlNetTransformationTool(Tool):
    default_stable_diffusion_checkpoint = "runwayml/stable-diffusion-v1-5"
    default_controlnet_checkpoint = "lllyasviel/control_v11p_sd15_canny"
    description = IMAGE_TRANSFORMATION_DESCRIPTION
    name = "image_transformer"
    inputs = ['image', 'text']
    outputs = ['image']

    def __init__(self, device=None, controlnet=None, stable_diffusion=None, **hub_kwargs) -> None:
        if not is_accelerate_available():
            raise ImportError("Accelerate should be installed in order to use tools.")
        if not is_vision_available():
            raise ImportError("Pillow should be installed in order to use the StableDiffusionTool.")

        super().__init__()

        self.stable_diffusion = self.default_stable_diffusion_checkpoint
        self.controlnet = self.default_controlnet_checkpoint

        self.device = device
        self.hub_kwargs = hub_kwargs

    def setup(self):
        if self.device is None:
            self.device = get_default_device()

        controlnet = ControlNetModel.from_pretrained(self.controlnet)
        self.pipeline = StableDiffusionControlNetPipeline.from_pretrained(self.stable_diffusion, controlnet=controlnet)
        self.pipeline.scheduler = UniPCMultistepScheduler.from_config(self.pipeline.scheduler.config)

        self.pipeline.to(self.device)
        if self.device.type == "cuda":
            self.pipeline.to(torch_dtype=torch.float16)

        self.is_initialized = True

    def __call__(self, image, prompt):
        if not self.is_initialized:
            self.setup()

        image = np.array(image)

        image = cv2.Canny(image, 100, 200)
        image = image[:, :, None]
        image = np.concatenate([image, image, image], axis=2)
        image = Image.fromarray(image)

        negative_prompt = "low quality, bad quality, deformed, low resolution"
        added_prompt = " , highest quality, highly realistic, very high resolution"

        return self.pipeline(
            prompt + added_prompt,
            image,
            negative_prompt=negative_prompt,
            num_inference_steps=30,
        ).images[0]