File size: 4,114 Bytes
74a2a96 |
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 |
import gradio as gr
import numpy as np
import random
import os
from PIL import Image
import spaces
import torch
from transformers import pipeline
from diffusers import StableDiffusionDepth2ImgPipeline
model_id_depth = "depth-anything/Depth-Anything-V2-Large-hf"
if torch.cuda.is_available():
pipe_depth = pipeline(task="depth-estimation", model=model_id_depth, device="cuda")
else:
pipe_depth = pipeline(task="depth-estimation", model=model_id_depth)
model_id_depth2image = "stabilityai/stable-diffusion-2-depth"
if torch.cuda.is_available():
pipe_depth2image = StableDiffusionDepth2ImgPipeline.from_pretrained(model_id_depth2image, torch_dtype=torch.float16).to("cuda")
else:
pipe_depth2image = StableDiffusionDepth2ImgPipeline.from_pretrained(model_id_depth2image)
max_seed = np.iinfo(np.int32).max
max_image_size = 1344
example_files = [os.path.join('assets/examples', filename) for filename in sorted(os.listdir('assets/examples'))]
@spaces.GPU
def infer(
init_image,
prompt,
negative_prompt,
seed,
randomize_seed,
width,
height,
guidance_scale,
num_inference_steps):
if randomize_seed:
seed = random.randint(0, max_seed)
init_image = Image.fromarray(np.uint8(init_image))
# generate depth
predicted_depth = pipe_depth(init_image)["predicted_depth"]
# generate image
image = pipe_depth2image(
prompt=prompt,
image=init_image,
depth_map=predicted_depth,
negative_prompt=negative_prompt,
guidance_scale=guidance_scale,
num_inference_steps=num_inference_steps,
height=height,
width=width,
generator=torch.Generator().manual_seed(seed)
).images[0]
return image, seed
with gr.Blocks() as demo:
gr.Markdown("# Demo [Depth2Image](https://huggingface.co/stabilityai/stable-diffusion-2-depth) with depth map estimated by [Depth Anything V2](https://huggingface.co/depth-anything/Depth-Anything-V2-Large-hf).")
prompt = gr.Text(
label="Prompt",
show_label=False,
max_lines=1,
placeholder="Enter your prompt",
container=False,
)
with gr.Row():
init_image = gr.Image(label="Input Image", type='numpy')
result = gr.Image(label="Result", show_label=False)
run_button = gr.Button("Run", scale=0)
with gr.Accordion("Advanced Settings", open=False):
negative_prompt = gr.Text(
label="Negative Prompt",
max_lines=1,
placeholder="Enter a negative prompt",
)
seed = gr.Slider(
label="Seed",
minimum=0,
maximum=max_seed,
step=1,
value=0,
)
randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
with gr.Row():
width = gr.Slider(
label="Width",
minimum=256,
maximum=max_image_size,
step=64,
value=1024,
)
height = gr.Slider(
label="Height",
minimum=256,
maximum=max_image_size,
step=64,
value=1024,
)
with gr.Row():
guidance_scale = gr.Slider(
label="Guidance scale",
minimum=0.0,
maximum=10.0,
step=0.1,
value=7.5,
)
num_inference_steps = gr.Slider(
label="Number of inference steps",
minimum=1,
maximum=50,
step=1,
value=5,
)
gr.on(
triggers=[run_button.click, prompt.submit, negative_prompt.submit],
fn=infer,
inputs=[init_image, prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
outputs=[result, seed]
)
examples = gr.Examples(
examples=example_files, inputs=[init_image], outputs=[depth_image_slider, gray_depth_file], fn=on_submit
)
demo.queue().launch() |