Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Abstract

Image-to-image translation is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image using a training set of aligned image pairs. However, for many tasks, paired training data will not be available. We present an approach for learning to translate an image from a source domain X to a target domain Y in the absence of paired examples. Our goal is to learn a mapping G: X → Y such that the distribution of images from G(X) is indistinguishable from the distribution Y using an adversarial loss. Because this mapping is highly under-constrained, we couple it with an inverse mapping F: Y → X and introduce a cycle consistency loss to push F(G(X)) ≈ X (and vice versa). Qualitative results are presented on several tasks where paired training data does not exist, including collection style transfer, object transfiguration, season transfer, photo enhancement, etc. Quantitative comparisons against several prior methods demonstrate the superiority of our approach.

Paper

arxiv 1703.10593, 2017.

Citation

Jun-Yan Zhu*, Taesung Park*, Phillip Isola, and Alexei A. Efros. "Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks", in arxiv, 2017.
(* indicates equal contributions) Bibtex

Code and Data: [Torch] [PyTorch]

Applications

Monet Paintings → Photos

Mapping Monet paintings to landscape photographs from Flickr:
[Best results], [Random training set results], [Random test set results]

Collection Style Transfer

Transferring input images into artistic styles of Monet, Van Gogh, Ukiyo-e, and Cezanne.
[Results on the author's personal photos]
[Random training set results], [Random test set results]

Object Transfiguration

Object transfiguration between horses and zebras:
[Best results], [Random training set results], [Random test set results]
Object transfiguration between apples and oranges:
[Best results], [Random training set results], [Random test set results]

Horse Video to Zebra Video

Season Transfer

Transferring seasons of Yosemite in the Flickr photos:
[Best results], [Random training set results], [Random test set results]

Photo Enhancement

iPhone photos → DSLR photos: generating photos with shallower depth of field.
[Best Results], [Random training set results], [Random test set results]

Experiments and comparisons

Comparison on Cityscapes: different methods for mapping labels ↔ photos trained on Cityscapes.
Comparison on Maps: different methods for mapping aerialphotos ↔ maps on Google Maps.
Facade results: CycleGAN for mapping labels ↔ facades on CMP Facades datasets.
Ablation studies: different variants of our method for mapping labels ↔ photos trained on Cityscapes.
Image reconstruction results: the reconstructed images F(G(x)) and G(F(y)) from various experiments.
Style transfer comparison: we compare our method with neural style transfer [Gatys et al. '15].
Identity mapping loss: the effect of the identity mapping loss on Monet to Photo.

Various Applications

Renoir Style: The movie Beauty and the Beast, 2017, rendered in the impressionism artist Renoir style

Failure Cases

Our model does not work well when a test image looks unusual compared to training images, as shown in the left figure.
See more typical failure cases [here]. On translation tasks that involve color and texture changes, like many of those reported above, the method often succeeds. We have also explored tasks that require geometric changes, with little success. For example, on the task of dog ↔ cat transfiguration, the learned translation degenerates to making minimal changes to the input. Handling more varied and extreme transformations, especially geometric changes, is an important problem for future work. We also observe a lingering gap between the results achievable with paired training data and those achieved by our unpaired method. In some cases, this gap may be very hard -- or even impossible,-- to close: for example, our method sometimes permutes the labels for tree and building in the output of the cityscapes photos → labels task. To resolve this ambiguity may require some form of weak semantic supervision. Integrating weak or semi-supervised data may lead to substantially more powerful translators, still at a fraction of the annotation cost of the fully-supervised systems.

Related Work

Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, Yoshua Bengio "Generative Adversarial Networks", in NIPS 2014.
Alec Radford, Luke Metz and Soumith Chintala "Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks", in ICLR 2016.
Jun-Yan Zhu, Philipp Krähenbühl, Eli Shechtman, and Alexei A. Efros. "Generative Visual Manipulation on the Natural Image Manifold", in ECCV 2016.
Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, and Alexei A. Efros. "Image-to-Image Translation with Conditional Adversarial Networks", in CVPR 2017.

Acknowledgement

We thank Aaron Hertzmann, Shiry Ginosar, Deepak Pathak, Bryan Russell, Eli Shechtman, Richard Zhang, and Tinghui Zhou for many helpful comments. This work was supported in part by NSF SMA-1514512, NSF IIS-1633310, a Google Research Award, Intel Corp, and hardware donations from NVIDIA. JYZ is supported by the Facebook Graduate Fellowship and TP is supported by the Samsung Scholarship. The photographs used in style transfer were taken by AE, mostly in France.