DREAMBOOTH: Train Your Own Style Like Midjourney On Stable Diffusion
84.4 هزار بار بازدید -
2 سال پیش
-
Dreambooth is Google’s new AI
Dreambooth is Google’s new AI and it allows you to train a stable diffusion model with your own pictures with better results than textual inversion. Dreambooth is originally based on Imagen text-to-image model and this technology makes it possible for you to insert any character (yourself, your friends, your family), object or animal you want into a stable diffusion model, all of that with a few images and in less than 1h. In this video, I will show you how you can train your own style similar to Midjourney using Dreambooth and GPU renting sites like Runpod or Vast ai, all of that for a few cents. I will also show how you can download a trained dreambooth models from the huggingface Stable Diffusion Dreambooth Concepts Library and then convert it into a CKPT file using the convert_diffusers_to_sd.py script that just came out a few days ago!
Did you manage to train a new style? Let me know in the comments!
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
SOCIAL MEDIA LINKS!
✨ Support my work on Patreon: Patreon: aitrepreneur
⚔️ Join the Discord server: https://bit.ly/aitdiscord
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Runpod: https://bit.ly/runpodAi
HUGGING FACE:
https://huggingface.co/sd-dreambooth-...
Midjourneyart style: https://mega.nz/folder/Z0xS1BpI#S40xU...
ddfusion style: https://mega.nz/folder/ph4GhKBQ#C7Yk2...
Converter script: https://gist.github.com/jachiam/8a5c0...
python .\convert_diffusers_to_sd.py --model_path .\disco-diffusion-style --checkpoint_path .\disco-diffusion-style\discodiffusion.ckpt
Download my repo by replacing the cell by:
dataset="style_ddim" #@param ["man_euler", "man_unsplash", "person_ddim", "woman_ddim", "blonde_woman"]
!git clone https://github.com/aitrepreneur/SD-Re...
!mkdir -p regularization_images/{dataset}
!mv -v SD-Regularization-Images-Style-Dreambooth/{dataset}/*.* regularization_images/{dataset}
./gdrive upload ./trained_models --recursive
#stablediffusion #dreambooth #stablediffusiontutorial
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
WATCH MY MOST POPULAR VIDEOS:
RECOMMENDED WATCHING - My "Stable Diffusion" Playlist:
►► https://bit.ly/stablediffusion
RECOMMENDED WATCHING - My "Tutorial" Playlist:
►► https://bit.ly/TuTPlaylist
Disclosure: Bear in mind that some of the links in this post are affiliate links and if you go through them to make a purchase I will earn a commission. Keep in mind that I link these companies and their products because of their quality and not because of the commission I receive from your purchases. The decision is yours, and whether or not you decide to buy something is completely up to you.
Did you manage to train a new style? Let me know in the comments!
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
SOCIAL MEDIA LINKS!
✨ Support my work on Patreon: Patreon: aitrepreneur
⚔️ Join the Discord server: https://bit.ly/aitdiscord
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Runpod: https://bit.ly/runpodAi
HUGGING FACE:
https://huggingface.co/sd-dreambooth-...
Midjourneyart style: https://mega.nz/folder/Z0xS1BpI#S40xU...
ddfusion style: https://mega.nz/folder/ph4GhKBQ#C7Yk2...
Converter script: https://gist.github.com/jachiam/8a5c0...
python .\convert_diffusers_to_sd.py --model_path .\disco-diffusion-style --checkpoint_path .\disco-diffusion-style\discodiffusion.ckpt
Download my repo by replacing the cell by:
dataset="style_ddim" #@param ["man_euler", "man_unsplash", "person_ddim", "woman_ddim", "blonde_woman"]
!git clone https://github.com/aitrepreneur/SD-Re...
!mkdir -p regularization_images/{dataset}
!mv -v SD-Regularization-Images-Style-Dreambooth/{dataset}/*.* regularization_images/{dataset}
./gdrive upload ./trained_models --recursive
#stablediffusion #dreambooth #stablediffusiontutorial
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
WATCH MY MOST POPULAR VIDEOS:
RECOMMENDED WATCHING - My "Stable Diffusion" Playlist:
►► https://bit.ly/stablediffusion
RECOMMENDED WATCHING - My "Tutorial" Playlist:
►► https://bit.ly/TuTPlaylist
Disclosure: Bear in mind that some of the links in this post are affiliate links and if you go through them to make a purchase I will earn a commission. Keep in mind that I link these companies and their products because of their quality and not because of the commission I receive from your purchases. The decision is yours, and whether or not you decide to buy something is completely up to you.
2 سال پیش
در تاریخ 1401/07/13 منتشر شده
است.
84,465
بـار بازدید شده