Spaces:
Running
Running
File size: 2,830 Bytes
0f63870 d098b29 0f63870 d098b29 0f63870 d098b29 0f63870 d098b29 0f63870 d098b29 0f63870 d098b29 0f63870 d098b29 0f63870 d098b29 0f63870 d098b29 0f63870 d098b29 0f63870 | 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 | import gradio as gr
import torch
import numpy as np
import random
from diffusers import StableDiffusionImg2ImgPipeline
from PIL import Image
dtype = torch.float16
device = "cuda" if torch.cuda.is_available() else "cpu"
MAX_SEED = np.iinfo(np.int32).max
MODEL_ID = "runwayml/stable-diffusion-v1-5"
STYLES = {
"Anime": {
"prompt": "anime style portrait, detailed face, vibrant colors, high quality, studio ghibli"
},
"Pixel Art": {
"prompt": "pixel art avatar, 16-bit style, retro game character, sharp pixels"
},
"Ghibli": {
"prompt": "ghibli studio style illustration, soft colors, dreamy atmosphere, watercolor"
},
}
print("Chargement du modèle...")
pipe = StableDiffusionImg2ImgPipeline.from_pretrained(
MODEL_ID,
torch_dtype=dtype,
safety_checker=None
)
pipe.to(device)
print("Modèle chargé !")
def generate_avatar(
input_image,
style_name,
strength,
seed,
randomize_seed,
progress=gr.Progress(track_tqdm=True)
):
if input_image is None:
raise gr.Error("Veuillez uploader une photo !")
if randomize_seed:
seed = random.randint(0, MAX_SEED)
generator = torch.Generator(device=device).manual_seed(seed)
prompt = STYLES[style_name]["prompt"]
input_image = input_image.resize((512, 512))
result = pipe(
prompt=prompt,
image=input_image,
strength=strength,
num_inference_steps=20,
guidance_scale=7.5,
generator=generator,
).images[0]
return (input_image, result), seed
with gr.Blocks() as demo:
gr.Markdown("# Avatar Generator — Transformez votre photo en personnage")
with gr.Row():
with gr.Column():
input_img = gr.Image(label="Votre photo", type="pil")
style = gr.Radio(
choices=list(STYLES.keys()),
label="Style d'avatar",
value="Anime"
)
strength = gr.Slider(
label="Intensité du style",
minimum=0.3, maximum=0.95,
step=0.05, value=0.7,
info="Bas = garde votre visage, Haut = stylisation forte"
)
with gr.Accordion("Paramètres avancés", open=False):
seed = gr.Slider(0, MAX_SEED, step=1, value=42, label="Seed")
randomize_seed = gr.Checkbox(label="Seed aléatoire", value=True)
btn = gr.Button("Générer mon avatar", variant="primary")
result_slider = gr.ImageSlider(label="Avant / Après", type="pil")
seed_out = gr.Number(label="Seed utilisé", visible=False)
btn.click(
fn=generate_avatar,
inputs=[input_img, style, strength, seed, randomize_seed],
outputs=[result_slider, seed_out]
)
if __name__ == "__main__":
demo.launch() |