Hi,

this repository has really helped me quite a lot in my research!!!
I uploaded our new work for unsupervised image to image translation, specifically for manifesting latent space constraints. see below for more info:
https://arxiv.org/abs/1901.04530

I would be grateful if you can add our work to the list of unsupervised image to image translation.

Thank you

https://github.com/Janspiry/Palette-Image-to-Image-Diffusion-Models (unofficial implementation)

Looking forword to your reply.

Hi,
this repository has helped it a lot during my research, thank you for opening and maintaining it.

We have uploaded our new work for unsupervised image to image translation, specifically for cross-domain translation (i.e. animals), by utilizing the semantics encoded in pre-trained network, e.g. VGG. Our translation is performed in a cascaded, deep-to-shallow, fashion, along the deep feature hierarchy, see here for more info.

I would be grateful if you can add our work to the list of unsupervised image to image translation.

Thank you

1. DRIT++: Diverse Image-to-Image Translation via Disentangled Representations
   https://arxiv.org/pdf/1905.01270.pdf
   extend drit to mult-domain task.
2. FUNIT: Few-Shot Unsupervised Image-to-Image Translation
   https://arxiv.org/pdf/1905.01723.pdf
   find general coding for class by few images.
3. Attribute Guided Unpaired Image-to-Image Translation with Semi-supervised Learning
   https://arxiv.org/pdf/1904.12428.pdf
   jointly solve multi-modal and multi-label tasks, introduce a semi-supervised training process(previous 'unsupervised' is actually unpaired), and some new tasks are proposed.

I think style transfer is a sub-class of image translation. How about you？

Hello, I am one of the authors of "GANHOPPER: Multi-Hop GAN for Unsupervised Image-to-Image Translation".

I believe this paper should be in the category "Unsupervised - General"

Also, the code was published at https://github.com/wallacemplira/ganhopper

Thank you for your work compiling this list!

arxiv: https://arxiv.org/abs/1903.10146
github: https://github.com/youyuge34/PI-REC

Can you add it? Thanks a lot!
Or maybe I could pull a request.

1. Multi-Channel Attention Selection GAN with Cascaded Semantic Guidance for Cross-View Image Translation (CVPR 2019); code: https://github.com/Ha0Tang/SelectionGAN
2. GestureGAN for Hand Gesture-to-Gesture Translation in the Wild (ACM MM 2018); code: https://github.com/Ha0Tang/GestureGAN
3. Attention-Guided Generative Adversarial Networks for Unsupervised Image-to-Image Translation (IJCNN 2019 oral); code: https://github.com/Ha0Tang/AttentionGAN

SingleGAN: Image-to-Image Translation by a Single-Generator Network using Multiple Generative Adversarial Learning

1. Image-to-Image Translation with Multi-Path Consistency Regularization
   https://arxiv.org/pdf/1905.12498.pdf
2. Mask-Guided Portrait Editing with Conditional GANs
   https://arxiv.org/pdf/1905.10346.pdf
3. Toward Learning a Unified Many-to-Many Mapping for Diverse Image Translation
   https://arxiv.org/pdf/1905.08766.pdf

And AGUIT's project: https://github.com/imlixinyang/AGUIT

https://eccv2018.org/openaccess/content_ECCV_2018/papers/Xinyuan_Chen_Attention-GAN_for_Object_ECCV_2018_paper.pdf

[1]Video-to-Video Synthesis https://tcwang0509.github.io/vid2vid/
[2] everybody dance now https://arxiv.org/abs/1808.07371

XGAN achieves translation from face to cartoon, which is a two domain task.
So I believe it should be in a same category as UNIT or CycleGAN.

CrossNet has now been published in WACV 2020