app.py

import torch
from diffusers import StableDiffusionPipeline, DDIMScheduler, AutoencoderKL
from PIL import Image
from ip_adapter.ip_adapter_faceid import IPAdapterFaceIDPlus
import cv2
from insightface.app import FaceAnalysis
from insightface.utils import face_align


base_model_path = "SG161222/Realistic_Vision_V4.0_noVAE"
vae_model_path = "stabilityai/sd-vae-ft-mse"
#image_encoder_path = "h94/IP-Adapter/models/image_encoder"
image_encoder_path = "image_encoder"
ip_ckpt = "IP-Adapter-FaceID/ip-adapter-faceid-plus_sd15.bin"


if torch.cuda.is_available():
    device = 'cuda'
    torch_dtype = torch.float16
else:
    device = 'cpu'
    torch_dtype = torch.float32
print(f'Using device: {device}')

noise_scheduler = DDIMScheduler(
    num_train_timesteps=1000,
    beta_start=0.00085,
    beta_end=0.012,
    beta_schedule="scaled_linear",
    clip_sample=False,
    set_alpha_to_one=False,
    steps_offset=1,
)
vae = AutoencoderKL.from_pretrained(vae_model_path).to(dtype=torch_dtype)
pipe = StableDiffusionPipeline.from_pretrained(
    base_model_path,
    torch_dtype=torch_dtype,
    scheduler=noise_scheduler,
    vae=vae,
    feature_extractor=None,
    safety_checker=None
)


ip_model = IPAdapterFaceIDPlus(pipe, image_encoder_path, ip_ckpt, device, num_tokens=4, torch_dtype=torch_dtype)



app = FaceAnalysis(name="buffalo_l", providers=['CUDAExecutionProvider', 'CPUExecutionProvider'])
app.prepare(ctx_id=0, det_size=(640, 640), det_thresh=0.2)

def generate_images(img_filepath, prompt, n_images=3,
                    negative_prompt="monochrome, lowres, bad anatomy, worst quality, low quality, blurry",
                    img_prompt_scale=0.5,
                    num_inference_steps=30,
                    seed=None):
    print(prompt)
    image = cv2.imread(img_filepath)
    faces = app.get(image)

    faceid_embeds = torch.from_numpy(faces[0].normed_embedding).unsqueeze(0)
    face_image = face_align.norm_crop(image, landmark=faces[0].kps, image_size=224) # you can also segment the face
    images = ip_model.generate(
        prompt=prompt, negative_prompt=negative_prompt, face_image=face_image, faceid_embeds=faceid_embeds,
        num_samples=n_images, width=512, height=512, num_inference_steps=num_inference_steps, seed=seed,
        scale=img_prompt_scale, # with scale=1 I get weird images
    )
    return [images, Image.fromarray(face_image[..., [2, 1, 0]])]

import gradio as gr
with gr.Blocks() as demo:
    gr.Markdown(
    """
    # IP-Adapter-FaceID-plus
    Generate images conditioned on a image prompt and a text prompt. Learn more here: https://huggingface.co/h94/IP-Adapter-FaceID
    """)
    with gr.Row():
        with gr.Column():
            demo_inputs = []
            demo_inputs.append(gr.Image(type='filepath', label='image prompt'))
            demo_inputs.append(gr.Textbox(label='text prompt', value='headshot of a man, green moss wall in the background'))
            demo_inputs.append(gr.Slider(maximum=3, minimum=1, value=3, step=1, label='number of images'))
            with gr.Accordion(label='Advanced options', open=False):
                demo_inputs.append(gr.Textbox(label='negative text prompt', value="monochrome, lowres, bad anatomy, worst quality, low quality, blurry"))
                demo_inputs.append(gr.Slider(maximum=1, minimum=0, value=0.5, step=0.05, label='image prompt scale'))
            btn = gr.Button("Generate")

        with gr.Column():
            demo_outputs = []
            demo_outputs.append(gr.Gallery(label='generated images'))
            demo_outputs.append(gr.Image(label='detected face', height=224, width=224))
    btn.click(generate_images, inputs=demo_inputs, outputs=demo_outputs)
    sample_prompts = [
        'headshot of a man, green moss wall in the background',
        'linkedin profile picture of a macdonalds worker',
        'LinkedIn profile picture of a beautiful man dressed in a suit, huge explosion in the background',
        ]
    gr.Examples(sample_prompts, inputs=demo_inputs[1], label='Sample prompts')

demo.launch(share=True, debug=True)