root
update readme
81171dc
metadata
language: zh
license: creativeml-openrail-m
tags:
  - stable-diffusion
  - stable-diffusion-diffusers
  - text-to-image
  - multilingual
  - English(En)
  - Chinese(Zh)
  - Spanish(Es)
  - French(Fr)
  - Russian(Ru)
  - Japanese(Ja)
  - Korean(Ko)
  - Arabic(Ar)
  - Italian(It)
  - diffusers
widget:
  - text: 一张<鸣人>男孩的照片
    example_title: 一张<鸣人>男孩的照片

This is a DreamBooth model finetuned from the multilingual text-to-image model AltDiffusion.

Dreambooth is one of the method of finetune the pretrained text-to-image model.Given as input just a few images of a subject, it learns to bind a unique identifier with that specific subject.

AltDiffusion which is a multilingual text-to-image model.It currently supports 9 languages and will support more in future. We can use the dreambooth to finetune the AltDiffusion,so that we can get a multilingual Dreambooth model which could supports 9 languages.

Here we give a example model finetuned use a dozen pictures of Uzumaki Naruto downloaded from web.The example code of inference is shown bellow.You can have a try and maybe train your own dreambooth.Hopes have fun!

Example code of inference

from diffusers import AltDiffusionPipeline, DPMSolverMultistepScheduler
import torch

pipe = AltDiffusionPipeline.from_pretrained("BAAI/DreamBooth-AltDiffusion")
pipe = pipe.to("cuda")

pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)

prompt = "一张<鸣人>男孩的照片"
image = pipe(prompt, num_inference_steps=25).images[0]
image.show()