from __future__ import annotations import pathlib import gradio as gr import torch import os import PIL import torchvision.transforms as T import torch.nn.functional as F import numpy as np import cv2 import matplotlib.pyplot as plt from typing import Any from transformers import ( CLIPTextModelWithProjection, CLIPVisionModelWithProjection, CLIPImageProcessor, CLIPTokenizer, ) from transformers import CLIPTokenizer from src.priors.lambda_prior_transformer import ( PriorTransformer, ) # original huggingface prior transformer without time conditioning from src.pipelines.pipeline_kandinsky_subject_prior import KandinskyPriorPipeline from diffusers import DiffusionPipeline from PIL import Image __device__ = "cpu" __dtype__ = torch.float32 if torch.cuda.is_available(): __device__ = "cuda" __dtype__ = torch.float16 class Model: def __init__(self): self.device = __device__ self.text_encoder = ( CLIPTextModelWithProjection.from_pretrained( "laion/CLIP-ViT-bigG-14-laion2B-39B-b160k", projection_dim=1280, torch_dtype=__dtype__, ) .eval() .requires_grad_(False) ).to(self.device) self.tokenizer = CLIPTokenizer.from_pretrained( "laion/CLIP-ViT-bigG-14-laion2B-39B-b160k", ) prior = PriorTransformer.from_pretrained( "ECLIPSE-Community/Lambda-ECLIPSE-Prior-v1.0", torch_dtype=__dtype__, ) self.pipe_prior = KandinskyPriorPipeline.from_pretrained( "kandinsky-community/kandinsky-2-2-prior", prior=prior, torch_dtype=__dtype__, ).to(self.device) self.pipe = DiffusionPipeline.from_pretrained( "kandinsky-community/kandinsky-2-2-decoder", torch_dtype=__dtype__ ).to(self.device) def inference(self, raw_data, seed): generator = torch.Generator(device="cuda").manual_seed(seed) image_emb, negative_image_emb = self.pipe_prior( raw_data=raw_data, generator=generator, ).to_tuple() image = self.pipe( image_embeds=image_emb, negative_image_embeds=negative_image_emb, num_inference_steps=50, guidance_scale=7.5, generator=generator, ).images[0] return image def run( self, image: dict[str, PIL.Image.Image], keyword: str, image2: dict[str, PIL.Image.Image], keyword2: str, text: str, seed: int, ): sub_imgs = [image["composite"]] sun_keywords = [keyword] if keyword2 and keyword2 != "no subject": sun_keywords.append(keyword2) if image2: sub_imgs.append(image2["composite"]) raw_data = { "prompt": text, "subject_images": sub_imgs, "subject_keywords": sun_keywords, } image = self.inference(raw_data, seed) return image def create_demo(): USAGE = """## To run the demo, you should: 1. Upload your image. 2. **Upload a masked subject image with white blankspace or whiten out manually using brush tool.** 3. Input a Keyword i.e. 'Dog' 4. For MultiSubject personalization, 4-1. Upload another image. 4-2. Input the Keyword i.e. 'Sunglasses' 3. Input proper text prompts, such as "A photo of Dog" or "A Dog wearing sunglasses", Please use the same keyword in the prompt. 4. Click the Run button. """ model = Model() with gr.Blocks() as demo: gr.HTML( """

λ-ECLIPSE: Multi-Concept Personalized Text-to-Image Diffusion Models by Leveraging CLIP Latent Space

Project Page | Paper

This demo is currently hosted on either a small GPU or CPU. We will soon provide high-end GPU support.

Please follow the instructions from here to run it locally: GitHub Inference Code

Open In Colab """ ) gr.Markdown(USAGE) with gr.Row(): with gr.Column(): with gr.Group(): gr.Markdown( "Upload your first masked subject image or mask out marginal space" ) image = gr.ImageEditor( label="Input", type="pil", brush=gr.Brush(colors=["#FFFFFF"], color_mode="fixed"), ) keyword = gr.Text( label="Keyword", placeholder='e.g. "Dog", "Goofie"', info="Keyword for first subject", ) gr.Markdown( "For Multi-Subject generation : Upload your second masked subject image or mask out marginal space" ) image2 = gr.ImageEditor( label="Input", type="pil", brush=gr.Brush(colors=["#FFFFFF"], color_mode="fixed"), ) keyword2 = gr.Text( label="Keyword", placeholder='e.g. "Sunglasses", "Grand Canyon"', info="Keyword for second subject", ) prompt = gr.Text( label="Prompt", placeholder='e.g. "A photo of dog", "A dog wearing sunglasses"', info="Keep the keywords used previously in the prompt", ) run_button = gr.Button("Run") with gr.Column(): result = gr.Image(label="Result") inputs = [ image, keyword, image2, keyword2, prompt, ] gr.Examples( examples=[ [ os.path.join(os.path.dirname(__file__), "./assets/luffy.jpg"), "luffy", os.path.join(os.path.dirname(__file__), "./assets/white.jpg"), "no subject", "luffy holding a sword", ], [ os.path.join(os.path.dirname(__file__), "./assets/luffy.jpg"), "luffy", os.path.join(os.path.dirname(__file__), "./assets/white.jpg"), "no subject", "luffy in the living room", ], [ os.path.join(os.path.dirname(__file__), "./assets/teapot.jpg"), "teapot", os.path.join(os.path.dirname(__file__), "./assets/white.jpg"), "no subject", "teapot on a cobblestone street", ], [ os.path.join(os.path.dirname(__file__), "./assets/trex.jpg"), "trex", os.path.join(os.path.dirname(__file__), "./assets/white.jpg"), "no subject", "trex near a river", ], [ os.path.join(os.path.dirname(__file__), "./assets/cat.png"), "cat", os.path.join( os.path.dirname(__file__), "./assets/blue_sunglasses.png" ), "glasses", "A cat wearing glasses on a snowy field", ], [ os.path.join(os.path.dirname(__file__), "./assets/statue.jpg"), "statue", os.path.join(os.path.dirname(__file__), "./assets/toilet.jpg"), "toilet", "statue sitting on a toilet", ], [ os.path.join(os.path.dirname(__file__), "./assets/teddy.jpg"), "teddy", os.path.join(os.path.dirname(__file__), "./assets/luffy_hat.jpg"), "hat", "a teddy wearing the hat at a beach", ], [ os.path.join(os.path.dirname(__file__), "./assets/chair.jpg"), "chair", os.path.join(os.path.dirname(__file__), "./assets/table.jpg"), "table", "a chair and table in living room", ], ], inputs=inputs, fn=model.run, outputs=result, ) run_button.click(fn=model.run, inputs=inputs, outputs=result) return demo if __name__ == "__main__": demo = create_demo() demo.queue(max_size=20).launch()