Official repository of "Improved Diffusion-based Image Translation using Asymmetric Gradient Guidance"

Environment

Pytorch 1.9.0, Python 3.9

$ conda create --name AGG python=3.9
$ conda activate AGG
$ pip install ftfy regex matplotlib lpips kornia opencv-python torch==1.9.0+cu111 torchvision==0.10.0+cu111 -f https://download.pytorch.org/whl/torch_stable.html
$ pip install color-matcher
$ pip install git+https://github.com/openai/CLIP.git

Model download

To generate images, please download the pre-trained diffusion model

imagenet 256x256 LINK

FFHQ 256x256 LINK

download the model into ./checkpoints folder

For face identity loss when using FFHQ pre-trained model, download pre-trained ArcFace model LINK

save the model into ./id_model

Inference

Please refer to the bash script

./run_demo.sh

For memory saving, we can use single CLIP model with --clip_models 'ViT-B/32'

Our source code heavily rely on DiffuseIT