File size: 3,279 Bytes
3db013e
74a2a96
b7e10c3
74a2a96
3db013e
 
74a2a96
3db013e
 
 
b7e10c3
 
 
fad088c
7810d79
 
74a2a96
 
 
b7e10c3
3db013e
 
 
 
 
 
74a2a96
 
 
 
b7e10c3
3db013e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
74a2a96
 
 
b7e10c3
6dba575
b7e10c3
6dba575
74a2a96
b7e10c3
6dba575
74a2a96
b7e10c3
 
74a2a96
b7e10c3
 
74a2a96
b7e10c3
 
 
74a2a96
 
 
b7e10c3
 
74a2a96
df258c6
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
import torch
import gradio as gr
from diffusers.utils import load_image
import spaces
# from panna.pipeline import PipelineDepth2ImageV2
from panna import Depth2Image, DepthAnythingV2

# model = PipelineDepth2ImageV2()
model_depth = DepthAnythingV2("depth-anything/Depth-Anything-V2-Large-hf", torch_dtype=torch.float32)
model_image = Depth2Image("stabilityai/stable-diffusion-2-depth")
title = ("# [Depth2Image](https://huggingface.co/stabilityai/stable-diffusion-2-depth) with [DepthAnythingV2](https://huggingface.co/depth-anything/Depth-Anything-V2-Large-hf)\n"
         "Depth2Image with depth map predicted by DepthAnything V2. The demo is part of [panna](https://github.com/abacws-abacus/panna) project.")
example_files = []
for n in range(1, 10):
    load_image(f"https://huggingface.co/spaces/depth-anything/Depth-Anything-V2/resolve/main/assets/examples/demo{n:0>2}.jpg").save(f"demo{n:0>2}.jpg")
    example_files.append(f"demo{n:0>2}.jpg")


@spaces.GPU
def infer(init_image, prompt, negative_prompt, seed, width, height, guidance_scale, num_inference_steps):
    depth = model_depth.image2depth([init_image], return_tensor=True)
    return model_image.text2image(
        [init_image],
        depth_maps=depth,
        prompt=[prompt],
        negative_prompt=[negative_prompt],
        guidance_scale=guidance_scale,
        num_inference_steps=num_inference_steps,
        height=height,
        width=width,
        seed=seed
    )[0]


# @spaces.GPU
# def infer(init_image, prompt, negative_prompt, seed, width, height, guidance_scale, num_inference_steps):
#     return model(
#         init_image,
#         prompt=prompt,
#         negative_prompt=negative_prompt,
#         guidance_scale=guidance_scale,
#         num_inference_steps=num_inference_steps,
#         height=height,
#         width=width,
#         seed=seed
#     )


with gr.Blocks() as demo:
    gr.Markdown(title)
    with gr.Row():
        prompt = gr.Text(label="Prompt", show_label=True, max_lines=1, placeholder="Enter your prompt", container=False)
        run_button = gr.Button("Run", scale=0)
    with gr.Row():
        init_image = gr.Image(label="Input Image", type='pil')
        result = gr.Image(label="Result")
    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=1_000_000, step=1, value=0)
        with gr.Row():
            width = gr.Slider(label="Width", minimum=256, maximum=1344, step=64, value=1024)
            height = gr.Slider(label="Height", minimum=256, maximum=1344, 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="Inference steps", minimum=1, maximum=50, step=1, value=50)
    examples = gr.Examples(examples=example_files, inputs=[init_image])
    gr.on(
        triggers=[run_button.click, prompt.submit, negative_prompt.submit],
        fn=infer,
        inputs=[init_image, prompt, negative_prompt, seed, width, height, guidance_scale, num_inference_steps],
        outputs=[result]
    )
demo.launch(server_name="0.0.0.0")