app.py

import gradio as gr
import numpy as np

import os
from PIL import Image
import requests
from io import BytesIO
import io
import base64
from loadimg import load_img

hf_token = os.environ.get("HF_TOKEN_API_DEMO") # we get it from a secret env variable, such that it's private
auth_headers = {"api_token": hf_token}

def convert_image_to_base64_string(img):
    buffer = io.BytesIO()
    img.save(buffer, format="PNG")  # You can choose the format (e.g., "JPEG", "PNG")
    # Encode the buffer in base64
    image_base64_string = base64.b64encode(buffer.getvalue()).decode('utf-8')
    return f",{image_base64_string}"
    
def download_image(url):
    response = requests.get(url)
    img_bytes = BytesIO(response.content)
    return Image.open(img_bytes).convert("RGB")

def api_call(image_base64_file):
    url = "https://engine.prod.bria-api.com/v1/tailored-gen/restyle_portrait"
    
    payload = {
    "id_image_file": image_base64_file,
    "tailored_model_id": 598,
    }
    response = requests.post(url, json=payload, headers=auth_headers)
    response = response.json()
    res_image = download_image(response["image_res"])
    
    return res_image


def predict(img_input):
    img = load_img(img_input)
    img = img.convert("RGB")  
    image_base64_file = convert_image_to_base64_string(img)
    
    gen_img = api_call(image_base64_file)
    
    return gen_img


css = '''
.gradio-container{max-width: 1100px !important}
#image_upload{min-height:400px}
#image_upload [data-testid="image"], #image_upload [data-testid="image"] > div{min-height: 400px}
#mask_radio .gr-form{background:transparent; border: none}
#word_mask{margin-top: .75em !important}
#word_mask textarea:disabled{opacity: 0.3}
.footer {margin-bottom: 45px;margin-top: 35px;text-align: center;border-bottom: 1px solid #e5e5e5}
.footer>p {font-size: .8rem; display: inline-block; padding: 0 10px;transform: translateY(10px);background: white}
.dark .footer {border-color: #303030}
.dark .footer>p {background: #0b0f19}
.acknowledgments h4{margin: 1.25em 0 .25em 0;font-weight: bold;font-size: 115%}
#image_upload .touch-none{display: flex}
@keyframes spin {
    from {
        transform: rotate(0deg);
    }
    to {
        transform: rotate(360deg);
    }
}
#share-btn-container {padding-left: 0.5rem !important; padding-right: 0.5rem !important; background-color: #000000; justify-content: center; align-items: center; border-radius: 9999px !important; max-width: 13rem; margin-left: auto;}
div#share-btn-container > div {flex-direction: row;background: black;align-items: center}
#share-btn-container:hover {background-color: #060606}
#share-btn {all: initial; color: #ffffff;font-weight: 600; cursor:pointer; font-family: 'IBM Plex Sans', sans-serif; margin-left: 0.5rem !important; padding-top: 0.5rem !important; padding-bottom: 0.5rem !important;right:0;}
#share-btn * {all: unset}
#share-btn-container div:nth-child(-n+2){width: auto !important;min-height: 0px !important;}
#share-btn-container .wrap {display: none !important}
#share-btn-container.hidden {display: none!important}
#prompt input{width: calc(100% - 160px);border-top-right-radius: 0px;border-bottom-right-radius: 0px;}
#run_button {
    width: 100%;
    height: 50px;  /* Set a fixed height for the button */
    display: flex;
    align-items: center;
    justify-content: center;
}
#output-img img, #image_upload img {
    object-fit: contain; /* Ensure aspect ratio is preserved */
    width: 100%;
    height: auto; /* Let height adjust automatically */
}
#prompt-container{margin-top:-18px;}
#prompt-container .form{border-top-left-radius: 0;border-top-right-radius: 0}
#image_upload{border-bottom-left-radius: 0px;border-bottom-right-radius: 0px}
'''

image_blocks = gr.Blocks(css=css, elem_id="total-container")
with image_blocks as demo:
    with gr.Column(elem_id="col-container"):
        gr.Markdown("## BRIA Restyle Portrait API")
        gr.HTML('''
          <p style="margin-bottom: 10px; font-size: 94%">
            This demo showcases the BRIA Restyle Portrait capability, which lets you transform the style of a portrait while preserving the person's facial features using a tailored model and a reference image.<br> ComfyUI Node is available <a href=https://github.com/Bria-AI/ComfyUI-BRIA-API/tree/main?tab=readme-ov-file#tailored-generation-nodes>here</a>
          </p>
        ''')

    with gr.Row():
        with gr.Column():
            image = gr.Image(label="Upload an image")
            with gr.Column():
                btn = gr.Button("Restyle!", elem_id="run_button")
        
        with gr.Column():
            image_out = gr.Image(label="Output", elem_id="output-img")

    # Button click will trigger the inpainting function (now with prompt included)
    btn.click(fn=predict, inputs=image, outputs=[image_out], api_name='run')

    gr.HTML(
        """
            <div class="footer">
                <p>Model by <a href="https://huggingface.co/diffusers" style="text-decoration: underline;" target="_blank">Diffusers</a> - Gradio Demo by 🤗 Hugging Face
                </p>
            </div>
        """
    )

image_blocks.queue(max_size=25, api_open=False).launch(show_api=False)