Pix2Pix
Tensorflow 2.0 Implementation of the paper Image-to-Image Translation using Conditional GANs by Philip Isola, Jun-Yan Zhu, Tinghui Zhou and Alexei A. ...
pix2pix generative-adversarial-networks conditional-gan tensorflow keras activation-functions mish computer-vision paper code arxiv:1611.07004 research

Pix2Pix

Binder PWC PWC PWC

Tensorflow 2.0 Implementation of the paper Image-to-Image Translation using Conditional GANs by Philip Isola, Jun-Yan Zhu, Tinghui Zhou and Alexei A. Efros.

Experiments with Standard Architecture

Experiment 1

Resource Credits: Trained on Nvidia Quadro M4000 provided by Paperspace Gradient.

Dataset: Facades

Result:

Experiment 1 Result

Experiment 2

Resource Credits: Trained on Nvidia Quadro P5000 provided by Paperspace Gradient.

Dataset: Maps

Result:

Experiment 2 Result

Experiment 3

Resource Credits: Trained on Nvidia Tesla V100 provided by DeepWrex Technologies.

Dataset: Cityscapes

Result:

Experiment 3 Result

Experiments with Mish Activation Function

[Experiment 1 Mish

Resource Credits: Trained on Nvidia Quadro P5000 provided by Paperspace Gradient.

Dataset: Facades

Generator Architecture:

  • The Generator is a Unet-Like model with skip connections between encoder and decoder.
  • Encoder Block is Convolution -> BatchNormalization -> Activation (Mish)
  • Decode Blocks is Conv2DTranspose -> BatchNormalization -> Dropout (optional) -> Activation (Mish)

Discriminator:

  • PatchGAN Discriminator
  • Discriminator Block is Convolution -> BatchNormalization -> Activation (Mish)

Result:

Experiment 1 Mish Result

Experiment 2 Mish

Resource Credits: Trained on Nvidia Tesla P100 provided by Google Colab.

Dataset: Facades

Generator Architecture:

  • The Generator is a Unet-Like model with skip connections between encoder and decoder.
  • Encoder Block is Convolution -> BatchNormalization -> Activation (Mish)
  • Decode Blocks is Conv2DTranspose -> BatchNormalization -> Dropout (optional) -> Activation (Mish)

Discriminator:

  • PatchGAN Discriminator
  • Discriminator Block is Convolution -> BatchNormalization -> Activation (ReLU)

Result:

Experiment 2 Mish Result

Experiment 3 Mish

Resource Credits: Trained on Nvidia Quadro P5000 provided by Paperspace Gradient.

Dataset: Facades

Generator Architecture:

  • The Generator is a Unet-Like model with skip connections between encoder and decoder.
  • Encoder Block is Convolution -> BatchNormalization -> Activation (Mish)
  • Decode Blocks is Conv2DTranspose -> BatchNormalization -> Dropout (optional) -> Activation (Mish) for the first three blocks are Conv2DTranspose -> BatchNormalization -> Dropout (optional) -> Activation (ReLU)

Discriminator:

  • PatchGAN Discriminator
  • Discriminator Block is Convolution -> BatchNormalization -> Activation (ReLU)

Result:

Experiment 3 Mish Result

Don't forget to tag @soumik12345 in your comment, otherwise they may not be notified.

Authors original post
Software Development Engineer at IBM || Deep Learning Researcher
Share this project
Similar projects
Image-to-Image Translation with Conditional Adversarial Networks
Tensorflow port of Image-to-Image Translation with Conditional Adversarial Nets
CycleGAN
Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks.
Pix2Pix with Tf-js
Implementation of web friendly ML models using TensorFlow.js. pix2pix, face segmentation, fast style transfer and many more ...
SimpleGAN
A Tensorflow-based framework to ease the training of generative models