import os from collections import OrderedDict import gradio as gr import shutil import uuid import torch from pathlib import Path from lib.utils.iimage import IImage from PIL import Image from lib import models from lib.methods import rasg, sd, sr from lib.utils import poisson_blend, image_from_url_text TMP_DIR = 'gradio_tmp' if Path(TMP_DIR).exists(): shutil.rmtree(TMP_DIR) Path(TMP_DIR).mkdir(exist_ok=True, parents=True) os.environ['GRADIO_TEMP_DIR'] = TMP_DIR on_huggingspace = os.environ.get("SPACE_AUTHOR_NAME") == "PAIR" negative_prompt_str = "text, bad anatomy, bad proportions, blurry, cropped, deformed, disfigured, duplicate, error, extra limbs, gross proportions, jpeg artifacts, long neck, low quality, lowres, malformed, morbid, mutated, mutilated, out of frame, ugly, worst quality" positive_prompt_str = "Full HD, 4K, high quality, high resolution" example_inputs = [ ['assets/examples/images/a40.jpg', 'medieval castle'], ['assets/examples/images/a4.jpg', 'parrot'], ['assets/examples/images/a65.jpg', 'hoodie'], ['assets/examples/images/a54.jpg', 'salad'], ['assets/examples/images/a51.jpg', 'space helmet'], ['assets/examples/images/a46.jpg', 'stack of books'], ['assets/examples/images/a19.jpg', 'antique greek vase'], ['assets/examples/images/a2.jpg', 'sunglasses'], ] thumbnails = [ 'https://lh3.googleusercontent.com/pw/ABLVV87bkFc_SRKrbXuk5BTp18dETNm18MLbjoJo6JvwbIkYtjZXrjU_H1dCJIP799OJjHTZmo19mYVyMCC1RLmwqzoZrgwQzfB-SCtxLa83IbXBQ23xzmKoZgsRlPztxNJD6gmXzFyatdLRzDxHIusBQLUz=w3580-h1150-s-no-gm', 'https://lh3.googleusercontent.com/pw/ABLVV85RWtrpTf1tMp2p3q37eg5DlFp5znifALK_JTjvxJua8UYMjytVoEy2GUW2cLXgBvQyYKg7GvrWXQ5hkdAsyih5Rf4rFnDq-JoiQYhVZHStCZLKxmeAlQna5ZwMPVTKG1TK63DH_OdK58gvSjWtF2ww=w3580-h1152-s-no-gm', 'https://lh3.googleusercontent.com/pw/ABLVV84dkaU6SQs9fyDjajpk1X9JkYp_zQBEnPVL67oi11_05U6-Ys5ydQpuny8GBQCMyVbFKxJ5unn9w__gmP9K0cKQ4_IVoT7Hvfmya71klDqSI7vu9Iy_5P2Il5-0giJFpumtffBA3kryn1xtJdR4vSA0=w2924-h1858-s-no-gm', 'https://lh3.googleusercontent.com/pw/ABLVV853ZyjvS4LvcPpVMY9BWz-232omt3-hgRiGcky_3ojE6WLKgtsrftsg1jSrUm2ccT_UOa279CulZy6fdnH_Xg1SunyRBxaRjOK0uxAkUFwb60rR1S4hI2MmhLV7KCi3tw1A-oiGi0f9JINyade-322A=w2622-h1858-s-no-gm', 'https://lh3.googleusercontent.com/pw/ABLVV86AJGUVGjb0i6CPg8zlJlWObNY0xdOzM1x5Bq9gKhP-ZWre5aaexRJDxQUO2gmJtRIyohD88FJDG_aVX2G5M0QOyGRWlZmx7tOVXLh-Kbesobxo9MfD-wqk9Ts9O8NUGtIwkWzo9SEs2opKdu83gB9F=w2528-h1858-s-no-gm', 'https://lh3.googleusercontent.com/pw/ABLVV87MplTciS7z-4i-eY3B3L0YhaK8UEQ3pTQD6W6uYVGR4hPD9u1WGEGyfg5ddqU-Bx2BrKskDhwxzF746cRhgFU5aPtbYA_-O7KfqXe9IsMxYCgUKxEHBm2ncqy64V-w-N8XOFgUMkAQqcuuNZ8Xapqp=w3580-h1186-s-no-gm', 'https://lh3.googleusercontent.com/pw/ABLVV877Esi6l2Kuw3akH5QBlmDAbWydZDZEEJqlZ_N-X7g33NQZU8nv_UKdAVETS7q23byTuldIAhW-q99zCycFB8Yfc-5e_WPNIM9icU0p3gd6DUVZR233ZNUtLca384MYGIhMGud9Y_Xed1I3PpiMhrpG=w2846-h1858-s-no-gm', 'https://lh3.googleusercontent.com/pw/ABLVV85hMQbSB6fCokdyut4ke7xTUqjERhuYygnj7T8IIA1k48e9GkaowDywPZzi5QJzZfj7wU3bgBHzjxop19qK1zOi5XDrjfXkn5bwj4MxicHa3TG-Rc-V-c1uyZVUyviyUlkGZ62FxuVROw2x0aGJIcr0=w3580-h1382-s-no-gm' ] example_previews = [ [thumbnails[0], 'Prompt: medieval castle'], [thumbnails[1], 'Prompt: parrot'], [thumbnails[2], 'Prompt: hoodie'], [thumbnails[3], 'Prompt: salad'], [thumbnails[4], 'Prompt: space helmet'], [thumbnails[5], 'Prompt: laptop'], [thumbnails[6], 'Prompt: antique greek vase'], [thumbnails[7], 'Prompt: sunglasses'], ] # Load models inpainting_models = OrderedDict([ ("Dreamshaper Inpainting V8", models.ds_inp.load_model()), ("Stable-Inpainting 2.0", models.sd2_inp.load_model()), ("Stable-Inpainting 1.5", models.sd15_inp.load_model()) ]) sr_model = models.sd2_sr.load_model(device='cuda:1') sam_predictor = models.sam.load_model(device='cuda:0') inp_model = inpainting_models[list(inpainting_models.keys())[0]] def set_model_from_name(inp_model_name): global inp_model print (f"Activating Inpaintng Model: {inp_model_name}") inp_model = inpainting_models[inp_model_name] def save_user_session(hr_image, hr_mask, lr_results, prompt, session_id=None): if session_id == '': session_id = str(uuid.uuid4()) tmp_dir = Path(TMP_DIR) session_dir = tmp_dir / session_id session_dir.mkdir(exist_ok=True, parents=True) hr_image.save(session_dir / 'hr_image.png') hr_mask.save(session_dir / 'hr_mask.png') lr_results_dir = session_dir / 'lr_results' if lr_results_dir.exists(): shutil.rmtree(lr_results_dir) lr_results_dir.mkdir(parents=True) for i, lr_result in enumerate(lr_results): lr_result.save(lr_results_dir / f'{i}.png') with open(session_dir / 'prompt.txt', 'w') as f: f.write(prompt) return session_id def recover_user_session(session_id): if session_id == '': return None, None, [], '' tmp_dir = Path(TMP_DIR) session_dir = tmp_dir / session_id lr_results_dir = session_dir / 'lr_results' hr_image = Image.open(session_dir / 'hr_image.png') hr_mask = Image.open(session_dir / 'hr_mask.png') lr_result_paths = list(lr_results_dir.glob('*.png')) gallery = [] for lr_result_path in sorted(lr_result_paths): gallery.append(Image.open(lr_result_path)) with open(session_dir / 'prompt.txt', "r") as f: prompt = f.read() return hr_image, hr_mask, gallery, prompt def rasg_run( use_painta, prompt, input, seed, eta, negative_prompt, positive_prompt, ddim_steps, guidance_scale=7.5, batch_size=1, session_id='' ): torch.cuda.empty_cache() seed = int(seed) batch_size = max(1, min(int(batch_size), 4)) image = IImage(input['image']).resize(512) mask = IImage(input['mask']).rgb().resize(512) method = ['rasg'] if use_painta: method.append('painta') method = '-'.join(method) inpainted_images = [] blended_images = [] for i in range(batch_size): seed = seed + i * 1000 inpainted_image = rasg.run( ddim=inp_model, method=method, prompt=prompt, image=image, mask=mask, seed=seed, eta=eta, negative_prompt=negative_prompt, positive_prompt=positive_prompt, num_steps=ddim_steps, guidance_scale=guidance_scale ).crop(image.size) blended_image = poisson_blend( orig_img=image.data[0], fake_img=inpainted_image.data[0], mask=mask.data[0], dilation=12 ) blended_images.append(blended_image) inpainted_images.append(inpainted_image.pil()) session_id = save_user_session( input['image'], input['mask'], inpainted_images, prompt, session_id=session_id) return blended_images, session_id def sd_run(use_painta, prompt, input, seed, eta, negative_prompt, positive_prompt, ddim_steps, guidance_scale=7.5, batch_size=1, session_id='' ): torch.cuda.empty_cache() seed = int(seed) batch_size = max(1, min(int(batch_size), 4)) image = IImage(input['image']).resize(512) mask = IImage(input['mask']).rgb().resize(512) method = ['default'] if use_painta: method.append('painta') method = '-'.join(method) inpainted_images = [] blended_images = [] for i in range(batch_size): seed = seed + i * 1000 inpainted_image = sd.run( ddim=inp_model, method=method, prompt=prompt, image=image, mask=mask, seed=seed, eta=eta, negative_prompt=negative_prompt, positive_prompt=positive_prompt, num_steps=ddim_steps, guidance_scale=guidance_scale ).crop(image.size) blended_image = poisson_blend( orig_img=image.data[0], fake_img=inpainted_image.data[0], mask=mask.data[0], dilation=12 ) blended_images.append(blended_image) inpainted_images.append(inpainted_image.pil()) session_id = save_user_session( input['image'], input['mask'], inpainted_images, prompt, session_id=session_id) return blended_images, session_id def upscale_run( ddim_steps, seed, use_sam_mask, session_id, img_index, negative_prompt='', positive_prompt=', high resolution professional photo' ): hr_image, hr_mask, gallery, prompt = recover_user_session(session_id) if len(gallery) == 0: return Image.open('./assets/sr_info.png') torch.cuda.empty_cache() seed = int(seed) img_index = int(img_index) img_index = 0 if img_index < 0 else img_index img_index = len(gallery) - 1 if img_index >= len(gallery) else img_index inpainted_image = gallery[img_index if img_index >= 0 else 0] output_image = sr.run( sr_model, sam_predictor, inpainted_image, hr_image, hr_mask, prompt=prompt + positive_prompt, noise_level=20, blend_trick=True, blend_output=True, negative_prompt=negative_prompt, seed=seed, use_sam_mask=use_sam_mask ) return output_image def switch_run(use_rasg, model_name, *args): set_model_from_name(model_name) if use_rasg: return rasg_run(*args) return sd_run(*args) with gr.Blocks(css='style.css') as demo: gr.HTML( """
For faster inference without waiting in queue, you may duplicate the space and upgrade to the suggested GPU in settings.