app.py

import cv2
import einops
import gradio as gr
import numpy as np
import torch

from pytorch_lightning import seed_everything
from util import resize_image, HWC3, apply_canny
from ldm.models.diffusion.ddim import DDIMSampler
from annotator.openpose import apply_openpose
from cldm.model import create_model, load_state_dict
from huggingface_hub import hf_hub_url, cached_download


REPO_ID = "lllyasviel/ControlNet"
scribble_checkpoint = "models/control_sd15_scribble.pth"
scribble_model = create_model('./models/cldm_v15.yaml').cpu()
scribble_model.load_state_dict(load_state_dict(cached_download(
    hf_hub_url(REPO_ID, scribble_checkpoint)
), location='cpu'))
scribble_model = scribble_model.cuda()
ddim_sampler_scribble = DDIMSampler(scribble_model)
save_memory = False

def process(input_image, prompt, input_control, num_samples, image_resolution, ddim_steps, scale, seed, eta, low_threshold, high_threshold):
    # TODO: Clean Function for single Task

    if input_control == "Scribble":
        return process_scribble(input_image, prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta)

def process_scribble(input_image, prompt, num_samples, image_resolution, ddim_steps, scale, seed, eta):

    with torch.no_grad():
        img = resize_image(HWC3(input_image), image_resolution)
        H, W, C = img.shape

        detected_map = np.zeros_like(img, dtype=np.uint8)
        detected_map[np.min(img, axis=2) < 127] = 255

        control = torch.from_numpy(detected_map.copy()).float().cuda() / 255.0
        control = torch.stack([control for _ in range(num_samples)], dim=0)
        control = einops.rearrange(control, 'b h w c -> b c h w').clone()

        seed_everything(seed)

        if save_memory:
            scribble_model.low_vram_shift(is_diffusing=False)

        cond = {"c_concat": [control], "c_crossattn": [scribble_model.get_learned_conditioning([prompt + ', ' + a_prompt] * num_samples)]}
        un_cond = {"c_concat": [control], "c_crossattn": [scribble_model.get_learned_conditioning([n_prompt] * num_samples)]}
        shape = (4, H // 8, W // 8)

        if save_memory:
            scribble_model.low_vram_shift(is_diffusing=False)
            
        samples, intermediates = ddim_sampler_scribble.sample(ddim_steps, num_samples,
                                                     shape, cond, verbose=False, eta=eta,
                                                     unconditional_guidance_scale=scale,
                                                     unconditional_conditioning=un_cond)

        if save_memory:
            scribble_model.low_vram_shift(is_diffusing=False)
                    
        x_samples = scribble_model.decode_first_stage(samples)
        x_samples = (einops.rearrange(x_samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().clip(0, 255).astype(np.uint8)

        results = [x_samples[i] for i in range(num_samples)]
    return [255 - detected_map] + results

    
def create_canvas(w, h):
    new_control_options = ["Interactive Scribble"]
    return np.zeros(shape=(h, w, 3), dtype=np.uint8) + 255


block = gr.Blocks().queue()
control_task_list = [
    "Scribble"
]

a_prompt = 'best quality, extremely detailed, architecture render, photorealistic, hyper realistic, surreal, dali, 3d rendering, render, 8k, 16k, extremely detailed, unreal engine, octane, maya'
n_prompt = 'longbody, lowres, bad anatomy, bad hands, missing fingers, pubic hair,extra digit, number, text, watermark, fewer digits, cropped, worst quality, low quality'

with block:
    gr.Markdown("## ControlNet - Architectural Sketch to Render Image")
    gr.HTML('''
     <p style="margin-bottom: 10px; font-size: 94%">
                Demo for ControlNet, Optimized for architectural sketch, based on <a href="https://github.com/lllyasviel/ControlNet" style="text-decoration: underline;" target="_blank">  lllyasviel ControlNet </a> implementation. 
              </p>
              ''')
    gr.HTML('''
     <p style="margin-bottom: 8px; font-size: 94%">
                HF Space created by Thaweewat Rugsujarit, If you have any suggestions or feedback, please feel free to contact me via <a href="https://www.linkedin.com/in/thaweewat-rugsujarit/" style="text-decoration: underline;" target="_blank">  Linkedin </a>. 
              </p>
              ''')
    with gr.Row():
        with gr.Column():
            input_image = gr.Image(source='upload', type="numpy")
            input_control = gr.Dropdown(control_task_list, value="Scribble", label="Task")
            prompt = gr.Textbox(label="Architectural Style")
            run_button = gr.Button(label="Run")
            
            with gr.Accordion("Advanced options", open=False):
                num_samples = gr.Slider(label="Images", minimum=1, maximum=12, value=1, step=1)
                image_resolution = gr.Slider(label="Image Resolution", minimum=256, maximum=768, value=512, step=256)
                low_threshold = gr.Slider(label="Canny low threshold", minimum=1, maximum=255, value=100, step=1)
                high_threshold = gr.Slider(label="Canny high threshold", minimum=1, maximum=255, value=200, step=1)
                ddim_steps = gr.Slider(label="Steps", minimum=1, maximum=100, value=20, step=1)
                scale = gr.Slider(label="Guidance Scale", minimum=0.1, maximum=30.0, value=9.0, step=0.1)
                seed = gr.Slider(label="Seed", minimum=0, maximum=2147483647, step=1, randomize=True)
                eta = gr.Slider(label="eta (DDIM)", minimum=0.0,maximum =1.0, value=0.0, step=0.1)

        with gr.Column():
            result_gallery = gr.Gallery(label='Output', show_label=False, elem_id="gallery").style(grid=2, height='auto')
    ips = [input_image, prompt, input_control, num_samples, image_resolution, ddim_steps, scale, seed, eta, low_threshold, high_threshold]
    run_button.click(fn=process, inputs=ips, outputs=[result_gallery])
    gr.Markdown("![visitor badge](https://visitor-badge.glitch.me/badge?page_id=Thaweewat.ControlNet-Architecture)")

block.launch(debug = True)