Skip to content

Unofficial implementation of Swapping Autoencoder for Deep Image Manipulation (https://arxiv.org/abs/2007.00653) in PyTorch

License

Notifications You must be signed in to change notification settings

rosinality/swapping-autoencoder-pytorch

Repository files navigation

swapping-autoencoder-pytorch

Unofficial implementation of Swapping Autoencoder for Deep Image Manipulation (https://arxiv.org/abs/2007.00653) in PyTorch

Usage

First create lmdb datasets:

python prepare_data.py --out LMDB_PATH --n_worker N_WORKER --size SIZE1,SIZE2,SIZE3,... DATASET_PATH

This will convert images to jpeg and pre-resizes it. This implementation does not use progressive growing, but you can create multiple resolution datasets using size arguments with comma separated lists, for the cases that you want to try another resolutions later.

Then you can train model in distributed settings

python -m torch.distributed.launch --nproc_per_node=N_GPU --master_port=PORT train.py --batch BATCH_SIZE LMDB_PATH

train.py supports Weights & Biases logging. If you want to use it, add --wandb arguments to the script.

Generate samples

You can test trained model using generate.py

python generate.py --ckpt [CHECKPOINT PATH] IMG1 IMG2 IMG3 ...

Samples

Generated sample image

About

Unofficial implementation of Swapping Autoencoder for Deep Image Manipulation (https://arxiv.org/abs/2007.00653) in PyTorch

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published