contrastive-unpaired-translation
contrastive-unpaired-translation copied to clipboard
taesungp
Contrastive unpaired image-to-image translation, faster and lighter training than cyclegan (ECCV 2020, in PyTorch)
After looking at cycle gan's horse2zebra translation i was experimenting with CUT but found it's results to be of poor quality. I am using the default settings for training.As such...
Hi all, Was wondering why the generator G was chosen as the network to extract intermediate feature maps for the contrastive loss. In other works on image translation, the discriminator...
Hello, great work! Do you perform any (differentiable) augmentations on the image patches before feeding them into the encoder-part of the generator, as is common in contrastive learning? I'm having...
--------------------------------------------------------------------------- NotImplementedError Traceback (most recent call last) [](https://yyxp5r04vsr-496ff2e9c6d22116-0-colab.googleusercontent.com/outputframe.html?vrz=colab-20220412-060046-RC00_441141632#) in () 10 # opt = TrainOptions().parse() # get training options 11 ---> 12 dataset = create_dataset() # create a dataset given...
![Uploading image.png…]() dataset [UnalignedDataset] was created model [CUTModel] was created The number of training images = 214 Setting up a new session... Question: Why does it keep getting stuck here...
This commit solves the #126 and the restriction of only support batch_size=1 in test.py. More, details: - A to B and B to A outputs are both available in the...
I am writing to let you know that the data_dependent_initialize uses set_input(data) prior to parallelize(), thus the whole model + data does not fit in one GPU and causes Memory...
Hi, I was wondering if this model can also used for paired image to image translation? If yes, what things need to be changed?
Hello Sir, When try training my-dataset (image size 512x512 3ch, png), I met batch-size error. My GPU's spec is GTX 1080 Ti. (12 G) I could train batch-size 1 only....
Hello, Thank you for the implementation of the Wasserstein GANs mode and GP loss! I followed the way proposed here: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/issues/439 It works for me for pix2pix and cycleGAN. I...
