NVIDIA / pix2pixHD

Synthesizing and manipulating 2048x1024 images with conditional GANs

4.9K Stars 1.0K Forks Last release: Not found Other 40 Commits 0 Releases

Available items

No Items, yet!

The developer of this repository has not created any items for sale yet. Need a bug fixed? Help with integration? A different license? Create a request here:


Project | Youtube | Paper

Pytorch implementation of our method for high-resolution (e.g. 2048x1024) photorealistic image-to-image translation. It can be used for turning semantic label maps into photo-realistic images or synthesizing portraits from face label maps.

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
Ting-Chun Wang1, Ming-Yu Liu1, Jun-Yan Zhu2, Andrew Tao1, Jan Kautz1, Bryan Catanzaro1
1NVIDIA Corporation, 2UC Berkeley
In CVPR 2018.

Image-to-image translation at 2k/1k resolution

  • Our label-to-streetview results

  • Interactive editing results

  • Additional streetview results

  • Label-to-face and interactive editing results

  • Our editing interface


  • Linux or macOS
  • Python 2 or 3
  • NVIDIA GPU (11G memory or larger) + CUDA cuDNN

Getting Started


  • Install PyTorch and dependencies from http://pytorch.org
  • Install python libraries dominate.
    pip install dominate
  • Clone this repo:
    git clone https://github.com/NVIDIA/pix2pixHD
    cd pix2pixHD


  • A few example Cityscapes test images are included in the
  • Please download the pre-trained Cityscapes model from here (google drive link), and put it under
  • Test the model (
    bash ./scripts/test_1024p.sh
    python test.py --name label2city_1024p --netG local --ngf 32 --resize_or_crop none
    The test results will be saved to a html file here:

More example scripts can be found in the



  • We use the Cityscapes dataset. To train a model on the full dataset, please download it from the official website (registration required). After downloading, please put it under the
    folder in the same way the example images are provided.


  • Train a model at 1024 x 512 resolution (
    bash ./scripts/train_512p.sh
    python train.py --name label2city_512p
  • To view training results, please checkout intermediate results in
    . If you have tensorflow installed, you can see tensorboard logs in
    by adding
    to the training scripts.

Multi-GPU training

  • Train a model using multiple GPUs (
    bash ./scripts/train_512p_multigpu.sh
    python train.py --name label2city_512p --batchSize 8 --gpu_ids 0,1,2,3,4,5,6,7
    Note: this is not tested and we trained our model using single GPU only. Please use at your own discretion.

Training with Automatic Mixed Precision (AMP) for faster speed

  • To train with mixed precision support, please first install apex from: https://github.com/NVIDIA/apex
  • You can then train the model by adding
    . For example,
    python -m torch.distributed.launch train.py --name label2city_512p --fp16
    In our test case, it trains about 80% faster with AMP on a Volta machine.

Training at full resolution

  • To train the images at full resolution (2048 x 1024) requires a GPU with 24G memory (
    bash ./scripts/train_1024p_24G.sh
    ), or 16G memory if using mixed precision (AMP).
  • If only GPUs with 12G memory are available, please use the 12G script (
    bash ./scripts/train_1024p_12G.sh
    ), which will crop the images during training. Performance is not guaranteed using this script.

Training with your own dataset

  • If you want to train with your own dataset, please generate label maps which are one-channel whose pixel values correspond to the object labels (i.e. 0,1,...,N-1, where N is the number of labels). This is because we need to generate one-hot vectors from the label maps. Please also specity
    --label_nc N
    during both training and testing.
  • If your input is not a label map, please just specify
    --label_nc 0
    which will directly use the RGB colors as input. The folders should then be named
    instead of
    , where the goal is to translate images from A to B.
  • If you don't have instance maps or don't want to use them, please specify
  • The default setting for preprocessing is
    , which will scale the width of all training images to
    (1024) while keeping the aspect ratio. If you want a different setting, please change it by using the
    option. For example,
    first resizes the image to have width
    and then does random cropping of size
    (opt.fineSize, opt.fineSize)
    skips the resizing step and only performs random cropping. If you don't want any preprocessing, please specify
    , which will do nothing other than making sure the image is divisible by 32.

More Training/Test Details

  • Flags: see
    for all the training flags; see
    for all the test flags.
  • Instance map: we take in both label maps and instance maps as input. If you don't want to use instance maps, please specify the flag


If you find this useful for your research, please use the following.

  title={High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs},
  author={Ting-Chun Wang and Ming-Yu Liu and Jun-Yan Zhu and Andrew Tao and Jan Kautz and Bryan Catanzaro},  
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},


This code borrows heavily from pytorch-CycleGAN-and-pix2pix.

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.