import argparse, os import cv2 import torch import numpy as np import torchvision from omegaconf import OmegaConf from PIL import Image from tqdm import tqdm, trange from itertools import islice from einops import rearrange from torchvision.utils import make_grid import time from pytorch_lightning import seed_everything from torch import autocast from contextlib import nullcontext from ldm.util import instantiate_from_config from ldm.models.diffusion.ddim import DDIMSampler from ldm.modules.diffusionmodules.openaimodel import clear_feature_dic,get_feature_dic from ldm.models.seg_module import Segmodule import numpy as np os.environ["CUDA_VISIBLE_DEVICES"] = "1" def chunk(it, size): it = iter(it) return iter(lambda: tuple(islice(it, size)), ()) def numpy_to_pil(images): """ Convert a numpy image or a batch of images to a PIL image. """ if images.ndim == 3: images = images[None, ...] images = (images * 255).round().astype("uint8") pil_images = [Image.fromarray(image) for image in images] return pil_images def load_model_from_config(config, ckpt, verbose=False): print(f"Loading model from {ckpt}") pl_sd = torch.load(ckpt, map_location="cpu") if "global_step" in pl_sd: print(f"Global Step: {pl_sd['global_step']}") sd = pl_sd["state_dict"] model = instantiate_from_config(config.model) m, u = model.load_state_dict(sd, strict=False) if len(m) > 0 and verbose: print("missing keys:") print(m) if len(u) > 0 and verbose: print("unexpected keys:") print(u) model.cuda() model.eval() return model def put_watermark(img, wm_encoder=None): if wm_encoder is not None: img = cv2.cvtColor(np.array(img), cv2.COLOR_RGB2BGR) img = wm_encoder.encode(img, 'dwtDct') img = Image.fromarray(img[:, :, ::-1]) return img def load_replacement(x): try: hwc = x.shape y = Image.open("assets/rick.jpeg").convert("RGB").resize((hwc[1], hwc[0])) y = (np.array(y)/255.0).astype(x.dtype) assert y.shape == x.shape return y except Exception: return x def plot_mask(img, masks, colors=None, alpha=0.8,indexlist=[0,1]) -> np.ndarray: """Visualize segmentation mask. Parameters ---------- img: numpy.ndarray Image with shape `(H, W, 3)`. masks: numpy.ndarray Binary images with shape `(N, H, W)`. colors: numpy.ndarray corlor for mask, shape `(N, 3)`. if None, generate random color for mask alpha: float, optional, default 0.5 Transparency of plotted mask Returns ------- numpy.ndarray The image plotted with segmentation masks, shape `(H, W, 3)` """ H,W= masks.shape[0],masks.shape[1] color_list=[[255,97,0],[128,42,42],[220,220,220],[255,153,18],[56,94,15],[127,255,212],[210,180,140],[221,160,221],[255,0,0],[255,128,0],[255,255,0],[128,255,0],[0,255,0],[0,255,128],[0,255,255],[0,128,255],[0,0,255],[128,0,255],[255,0,255],[255,0,128]]*6 final_color_list=[np.array([[i]*512]*512) for i in color_list] background=np.ones(img.shape)*255 count=0 colors=final_color_list[indexlist[count]] for mask, color in zip(masks, colors): color=final_color_list[indexlist[count]] mask = np.stack([mask, mask, mask], -1) img = np.where(mask, img * (1 - alpha) + color * alpha,background*0.4+img*0.6 ) count+=1 return img.astype(np.uint8) def main(): parser = argparse.ArgumentParser() parser.add_argument( "--prompt", type=str, nargs="?", default="a photo of a lion on a mountain top at sunset", help="the prompt to render" ) parser.add_argument( "--category", type=str, nargs="?", default="lion", help="the category to ground" ) parser.add_argument( "--outdir", type=str, nargs="?", help="dir to write results to", default="outputs/txt2img-samples" ) parser.add_argument( "--skip_grid", action='store_true', help="do not save a grid, only individual samples. Helpful when evaluating lots of samples", ) parser.add_argument( "--skip_save", action='store_true', help="do not save individual samples. For speed measurements.", ) parser.add_argument( "--ddim_steps", type=int, default=50, help="number of ddim sampling steps", ) parser.add_argument( "--plms", action='store_true', help="use plms sampling", ) parser.add_argument( "--laion400m", action='store_true', help="uses the LAION400M model", ) parser.add_argument( "--fixed_code", action='store_true', help="if enabled, uses the same starting code across samples ", ) parser.add_argument( "--ddim_eta", type=float, default=0.0, help="ddim eta (eta=0.0 corresponds to deterministic sampling", ) parser.add_argument( "--n_iter", type=int, default=1, help="sample this often", ) parser.add_argument( "--H", type=int, default=512, help="image height, in pixel space", ) parser.add_argument( "--W", type=int, default=512, help="image width, in pixel space", ) parser.add_argument( "--C", type=int, default=4, help="latent channels", ) parser.add_argument( "--f", type=int, default=8, help="downsampling factor", ) parser.add_argument( "--n_samples", type=int, default=1, help="how many samples to produce for each given prompt. A.k.a. batch size", ) parser.add_argument( "--n_rows", type=int, default=0, help="rows in the grid (default: n_samples)", ) parser.add_argument( "--scale", type=float, default=7.5, help="unconditional guidance scale: eps = eps(x, empty) + scale * (eps(x, cond) - eps(x, empty))", ) parser.add_argument( "--from-file", type=str, help="if specified, load prompts from this file", ) parser.add_argument( "--config", type=str, default="configs/stable-diffusion/v1-inference.yaml", help="path to config which constructs model", ) parser.add_argument( "--sd_ckpt", type=str, default="stable_diffusion.ckpt", help="path to checkpoint of stable diffusion model", ) parser.add_argument( "--grounding_ckpt", type=str, default="grounding_module.pth", help="path to checkpoint of grounding module", ) parser.add_argument( "--seed", type=int, default=42, help="the seed (for reproducible sampling)", ) parser.add_argument( "--precision", type=str, help="evaluate at this precision", choices=["full", "autocast"], default="autocast" ) opt = parser.parse_args() if opt.laion400m: print("Falling back to LAION 400M model...") opt.config = "configs/latent-diffusion/txt2img-1p4B-eval.yaml" opt.ckpt = "models/ldm/text2img-large/model.ckpt" opt.outdir = "outputs/txt2img-samples-laion400m" seed_everything(opt.seed) tic = time.time() config = OmegaConf.load(f"{opt.config}") model = load_model_from_config(config, f"{opt.sd_ckpt}") device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu") model = model.to(device) toc = time.time() seg_module=Segmodule().to(device) seg_module.load_state_dict(torch.load(opt.grounding_ckpt, map_location="cpu"), strict=True) print('load time:',toc-tic) sampler = DDIMSampler(model) os.makedirs(opt.outdir, exist_ok=True) outpath = opt.outdir batch_size = opt.n_samples precision_scope = autocast if opt.precision=="autocast" else nullcontext with torch.no_grad(): with precision_scope("cuda"): with model.ema_scope(): prompt = opt.prompt text = opt.category trainclass = text if not opt.from_file: assert prompt is not None data = [batch_size * [prompt]] else: print(f"reading prompts from {opt.from_file}") with open(opt.from_file, "r") as f: data = f.read().splitlines() data = list(chunk(data, batch_size)) sample_path = os.path.join(outpath, "samples") os.makedirs(sample_path, exist_ok=True) start_code = None if opt.fixed_code: print('start_code') start_code = torch.randn([opt.n_samples, opt.C, opt.H // opt.f, opt.W // opt.f], device=device) for n in trange(opt.n_iter, desc="Sampling"): for prompts in tqdm(data, desc="data"): clear_feature_dic() uc = None if opt.scale != 1.0: uc = model.get_learned_conditioning(batch_size * [""]) if isinstance(prompts, tuple): prompts = list(prompts) c = model.get_learned_conditioning(prompts) shape = [opt.C, opt.H // opt.f, opt.W // opt.f] samples_ddim,_, _ = sampler.sample(S=opt.ddim_steps, conditioning=c, batch_size=opt.n_samples, shape=shape, verbose=False, unconditional_guidance_scale=opt.scale, unconditional_conditioning=uc, eta=opt.ddim_eta, x_T=start_code) x_samples_ddim = model.decode_first_stage(samples_ddim) diffusion_features = get_feature_dic() x_sample = torch.clamp((x_samples_ddim[0] + 1.0) / 2.0, min=0.0, max=1.0) x_sample = 255. * rearrange(x_sample.cpu().numpy(), 'c h w -> h w c') Image.fromarray(x_sample.astype(np.uint8)).save("demo/demo.png") img = x_sample.astype(np.uint8) class_name = trainclass query_text ="a "+prompt.split()[1]+" of a "+class_name c_split = model.cond_stage_model.tokenizer.tokenize(query_text) sen_text_embedding = model.get_learned_conditioning(query_text) class_embedding = sen_text_embedding[:, 5:len(c_split)+1, :] if class_embedding.size()[1] > 1: class_embedding = torch.unsqueeze(class_embedding.mean(1), 1) text_embedding = class_embedding text_embedding = text_embedding.repeat(batch_size, 1, 1) pred_seg_total = seg_module(diffusion_features, text_embedding) pred_seg = torch.unsqueeze(pred_seg_total[0,0,:,:], 0).unsqueeze(0) label_pred_prob = torch.sigmoid(pred_seg) label_pred_mask = torch.zeros_like(label_pred_prob, dtype=torch.float32) label_pred_mask[label_pred_prob > 0.5] = 1 annotation_pred = label_pred_mask[0][0].cpu() mask = annotation_pred.numpy() mask = np.expand_dims(mask, 0) done_image_mask = plot_mask(img, mask, alpha=0.9, indexlist=[0]) cv2.imwrite(os.path.join("demo/demo_mask.png"), done_image_mask) torchvision.utils.save_image(annotation_pred, os.path.join("demo/demo_segresult.png"), normalize=True, scale_each=True) if __name__ == "__main__": main()