StarGAN - Official PyTorch Implementation (CVPR 2018)

Last update: Jan 04, 2023

Overview

StarGAN - Official PyTorch Implementation

***** New: StarGAN v2 is available at https://github.com/clovaai/stargan-v2 *****

This repository provides the official PyTorch implementation of the following paper:

StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation
Yunjey Choi^1,2, Minje Choi^1,2, Munyoung Kim^2,3, Jung-Woo Ha², Sung Kim^2,4, Jaegul Choo^1,2
¹Korea University, ²Clova AI Research, NAVER Corp.
³The College of New Jersey, ⁴Hong Kong University of Science and Technology
https://arxiv.org/abs/1711.09020

Abstract: Recent studies have shown remarkable success in image-to-image translation for two domains. However, existing approaches have limited scalability and robustness in handling more than two domains, since different models should be built independently for every pair of image domains. To address this limitation, we propose StarGAN, a novel and scalable approach that can perform image-to-image translations for multiple domains using only a single model. Such a unified model architecture of StarGAN allows simultaneous training of multiple datasets with different domains within a single network. This leads to StarGAN's superior quality of translated images compared to existing models as well as the novel capability of flexibly translating an input image to any desired target domain. We empirically demonstrate the effectiveness of our approach on a facial attribute transfer and a facial expression synthesis tasks.

Dependencies

Python 3.5+
PyTorch 0.4.0+
TensorFlow 1.3+ (optional for tensorboard)

Downloading datasets

To download the CelebA dataset:

git clone https://github.com/yunjey/StarGAN.git
cd StarGAN/
bash download.sh celeba

To download the RaFD dataset, you must request access to the dataset from the Radboud Faces Database website. Then, you need to create a folder structure as described here.

Training networks

To train StarGAN on CelebA, run the training script below. See here for a list of selectable attributes in the CelebA dataset. If you change the selected_attrs argument, you should also change the c_dim argument accordingly.

# Train StarGAN using the CelebA dataset
python main.py --mode train --dataset CelebA --image_size 128 --c_dim 5 \
               --sample_dir stargan_celeba/samples --log_dir stargan_celeba/logs \
               --model_save_dir stargan_celeba/models --result_dir stargan_celeba/results \
               --selected_attrs Black_Hair Blond_Hair Brown_Hair Male Young

# Test StarGAN using the CelebA dataset
python main.py --mode test --dataset CelebA --image_size 128 --c_dim 5 \
               --sample_dir stargan_celeba/samples --log_dir stargan_celeba/logs \
               --model_save_dir stargan_celeba/models --result_dir stargan_celeba/results \
               --selected_attrs Black_Hair Blond_Hair Brown_Hair Male Young

To train StarGAN on RaFD:

# Train StarGAN using the RaFD dataset
python main.py --mode train --dataset RaFD --image_size 128 \
               --c_dim 8 --rafd_image_dir data/RaFD/train \
               --sample_dir stargan_rafd/samples --log_dir stargan_rafd/logs \
               --model_save_dir stargan_rafd/models --result_dir stargan_rafd/results

# Test StarGAN using the RaFD dataset
python main.py --mode test --dataset RaFD --image_size 128 \
               --c_dim 8 --rafd_image_dir data/RaFD/test \
               --sample_dir stargan_rafd/samples --log_dir stargan_rafd/logs \
               --model_save_dir stargan_rafd/models --result_dir stargan_rafd/results

To train StarGAN on both CelebA and RafD:

# Train StarGAN using both CelebA and RaFD datasets
python main.py --mode=train --dataset Both --image_size 256 --c_dim 5 --c2_dim 8 \
               --sample_dir stargan_both/samples --log_dir stargan_both/logs \
               --model_save_dir stargan_both/models --result_dir stargan_both/results

# Test StarGAN using both CelebA and RaFD datasets
python main.py --mode test --dataset Both --image_size 256 --c_dim 5 --c2_dim 8 \
               --sample_dir stargan_both/samples --log_dir stargan_both/logs \
               --model_save_dir stargan_both/models --result_dir stargan_both/results

To train StarGAN on your own dataset, create a folder structure in the same format as RaFD and run the command:

# Train StarGAN on custom datasets
python main.py --mode train --dataset RaFD --rafd_crop_size CROP_SIZE --image_size IMG_SIZE \
               --c_dim LABEL_DIM --rafd_image_dir TRAIN_IMG_DIR \
               --sample_dir stargan_custom/samples --log_dir stargan_custom/logs \
               --model_save_dir stargan_custom/models --result_dir stargan_custom/results

# Test StarGAN on custom datasets
python main.py --mode test --dataset RaFD --rafd_crop_size CROP_SIZE --image_size IMG_SIZE \
               --c_dim LABEL_DIM --rafd_image_dir TEST_IMG_DIR \
               --sample_dir stargan_custom/samples --log_dir stargan_custom/logs \
               --model_save_dir stargan_custom/models --result_dir stargan_custom/results

Using pre-trained networks

To download a pre-trained model checkpoint, run the script below. The pre-trained model checkpoint will be downloaded and saved into ./stargan_celeba_128/models directory.

$ bash download.sh pretrained-celeba-128x128

To translate images using the pre-trained model, run the evaluation script below. The translated images will be saved into ./stargan_celeba_128/results directory.

$ python main.py --mode test --dataset CelebA --image_size 128 --c_dim 5 \
                 --selected_attrs Black_Hair Blond_Hair Brown_Hair Male Young \
                 --model_save_dir='stargan_celeba_128/models' \
                 --result_dir='stargan_celeba_128/results'

Citation

If you find this work useful for your research, please cite our paper:

@inproceedings{choi2018stargan,
author={Yunjey Choi and Minje Choi and Munyoung Kim and Jung-Woo Ha and Sunghun Kim and Jaegul Choo},
title={StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
year={2018}
}

Acknowledgements

This work was mainly done while the first author did a research internship at Clova AI Research, NAVER. We thank all the researchers at NAVER, especially Donghyun Kwak, for insightful discussions.

StarGAN - Official PyTorch Implementation (CVPR 2018)

Related tags

Overview

StarGAN - Official PyTorch Implementation

Dependencies

Downloading datasets

Training networks

Using pre-trained networks

Citation

Acknowledgements

Owner

Yunjey Choi

[Arxiv preprint] Causality-inspired Single-source Domain Generalization for Medical Image Segmentation (code&data-processing pipeline)

Code of paper "CDFI: Compression-Driven Network Design for Frame Interpolation", CVPR 2021

Model Serving Made Easy

Official Pytorch implementation of the paper "MotionCLIP: Exposing Human Motion Generation to CLIP Space"

Object detection using yolo-tiny model and opencv used as backend

PyTorch code accompanying our paper on Maximum Entropy Generators for Energy-Based Models

An unofficial personal implementation of UM-Adapt, specifically to tackle joint estimation of panoptic segmentation and depth prediction for autonomous driving datasets.

3D Human Pose Machines with Self-supervised Learning

Code for BMVC2021 "MOS: A Low Latency and Lightweight Framework for Face Detection, Landmark Localization, and Head Pose Estimation"

🤗 Transformers: State-of-the-art Natural Language Processing for Pytorch, TensorFlow, and JAX.

Differentiable Surface Triangulation

This repository attempts to replicate the SqueezeNet architecture and implement the same on an image classification task.

👨‍💻 run nanosaur in simulation with Gazebo/Ingnition

ML-Ensemble – high performance ensemble learning

Finite Element Analysis

Pytorch and Keras Implementations of Hyperspectral Image Classification -- Traditional to Deep Models: A Survey for Future Prospects.

Powerful unsupervised domain adaptation method for dense retrieval.

Hardware-accelerated DNN model inference ROS2 packages using NVIDIA Triton/TensorRT for both Jetson and x86_64 with CUDA-capable GPU

The NEOSSat is a dual-mission microsatellite designed to detect potentially hazardous Earth-orbit-crossing asteroids and track objects that reside in deep space

Official repository of my book: "Deep Learning with PyTorch Step-by-Step: A Beginner's Guide"