Skip to content

[ACCV 2024] Official PyTorch implementation of "Diffusion Model Compression for Image-to-Image Translation"

License

Notifications You must be signed in to change notification settings

KIMGEONUNG/ID-compression

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ID-Compression:
Diffusion Model Compression for Image-to-Image Translation

Geonung Kim  Beomsu Kim  Eunhyeok Park  Sunghyun Cho
POSTECH
ACCV 2024

teaser

🔥 Update

  • [2024.10.09] Depth-skip compression code is released
  • [2024.10.08] The repository is created.

🔧 Install Environment

Clone Repository

git clone [email protected]:KIMGEONUNG/ID-compression.git

Conda Environment

We offer three applications using our compression method: InstructPix2Pix for image editing, StableSR for image restoration, and ControlNet for structure-guided image synthesis. Please refer to the original application repository to install the conda environments.

Checkpoints

The checkpoint paths for each task are as follows. Please refer to the original repository to download the checkpoints.

# For InstructPix2Pix
InstructPix2Pix/ckpts
└── instruct-pix2pix-00-22000.ckpt

# For StableSR
StableSR/ckpts
├── stablesr_000117.ckpt
└── vqgan_cfw_00011.ckpt

# For ControlNet
ControlNet/models
└── control_sd15_canny.pth 

⌨️ Quick Start

InstructPix2Pix

cd InstructPix2Pix
python depth-skip.py --depth 9

StableSR

cd StableSR
python depth-skip.py --depth 9

ControlNet

cd ControlNet
python depth-skip.py --depth 9

☕️ Acknowledgment

About

[ACCV 2024] Official PyTorch implementation of "Diffusion Model Compression for Image-to-Image Translation"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published