Skip to content

enochkan/awesome-gans-and-deepfakes

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

40 Commits
 
 

Repository files navigation

Awesome-GANS-and-Deepfakes

Awesome Maintenance made-with-Markdown Documentation Status

A curated list of GAN & Deepfake papers and repositories. ✔️ means implementation is available.

GANs

Tl;dr GANs containg two competing neural networks which iteratively generate new data with the same statistics as the training set.

Unconditional GANs

  • ✔️ Vanilla GAN: Generative Adversarial Networks, [paper], [github]
  • ✔️ DCGAN: Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, [paper], [github]
  • ✔️ WGAN: Wasserstein GAN, [paper], [github]
  • ✔️ WGAN-GP: Improved Training of Wasserstein GANs, [paper], [github]
  • ✔️ RGAN: The relativistic discriminator: a key element missing from standard GAN, [paper], [github]
  • ✔️ BGAN: Boundary-Seeking Generative Adversarial Networks, [paper], [github]
  • ✔️ ClusterGAN: Latent Space Clustering in Generative Adversarial Networks, [paper], [github]

Conditional GANs

  • ✔️ CGAN: Conditional Generative Adversarial Nets, [paper], [github]
  • ✔️ ACGAN: Conditional Image Synthesis With Auxiliary Classifier GANs, [paper], [github]
  • ✔️ CCGAN: Semi-Supervised Learning with Context-Conditional Generative Adversarial Networks, [paper], [github]

Image-to-Image Translation

  • ✔️ CycleGAN: Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks, [paper], [github]
  • ✔️ StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation, [paper], [github]
  • ✔️ Pix2Pix: Image-to-Image Translation with Conditional Adversarial Nets, [paper], [github]
  • ✔️ DualGAN: Unsupervised Dual Learning for Image-to-Image Translation, [paper], [github]
  • ✔️ BicycleGAN: Toward Multimodal Image-to-Image Translation, [paper], [github]

Volumetric (3D) Generation

  • ✔️ 3DGAN: Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling, [paper], [github]
  • ✔️ Inverse Graphics GAN: Inverse Graphics GAN - Learning to Generate 3D Shapes from Unstructured 2D Data, [paper], [github]

Applications using GANs

Anime generator

  • Towards the Automatic Anime Characters Creation with Generative Adversarial Networks, [paper]
  • ✔️ [Project] Keras-GAN-Animeface-Character, [github]

Interactive Image generation

  • ✔️ Generative Visual Manipulation on the Natural Image Manifold, [paper], [github]
  • ✔️ Neural Photo Editing with Introspective Adversarial Networks, [paper], [github]

3D Object generation

  • 3D Shape Induction from 2D Views of Multiple Objects, [paper]
  • ✔️ Parametric 3D Exploration with Stacked Adversarial Networks, [github], [youtube]
  • ✔️ Fully Convolutional Refined Auto-Encoding Generative Adversarial Networks for 3D Multi Object Scenes, [github], [blog]

Super-resolution

  • ✔️ Image super-resolution through deep learning, [github]
  • ✔️ Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network, [paper], [github]
  • High-Quality Face Image Super-Resolution Using Conditional Generative Adversarial Networks, [paper]
  • ✔️ Analyzing Perception-Distortion Tradeoff using Enhanced Perceptual Super-resolution Network, [paper], [github]
  • ✔️ ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks, [paper], [github]
  • ✔️ MUNIT: Multimodal Unsupervised Image-to-Image Translation, [paper], [github]
  • ✔️ SRGAN: Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network, [paper], [github]

Image Inpainting (hole filling)

Medical Image Segmentation

  • ✔️ Vox2Vox: 3D-GAN for Brain Tumor Segmentation, [paper], [github]
  • SegAN: Adversarial Network with Multi-scale L1 Loss for Medical Image Segmentation, [paper]
  • Generative Adversarial Neural Networks for Pigmented and Non-Pigmented Skin Lesions Detection in Clinical Images, [paper]

Deepfakes

Tl;dr Deepfakes are fake videos or audio recordings that look and sound just like the real thing. Watch this video of Obama speaking... or was that really him?

CNN-based Face-swapping

  • ✔️ Fast Face-swap Using Convolutional Neural Networks, [paper], [github]
  • ✔️ DeepFaceLab: A simple, flexible and extensible face swapping framework, [paper], [github]

GAN-based Face-swapping

  • ✔️ Fewshot Face Translation GAN, [github]
  • Faceswap-GAN, [github]
  • ✔️ AttGAN: Facial Attribute Editing by Only Changing What You Want, [paper], [github]
  • MulGAN: Facial Attribute Editing by Exemplar, [paper]
  • ✔️ MaskGAN: Towards Diverse and Interactive Facial Image Manipulation, [paper], [github]
  • ✔️ StarGAN v2: Diverse Image Synthesis for Multiple Domains, [paper], [github]
  • ✔️ FSGAN: Subject Agnostic Face Swapping and Reenactment, [paper], [github]

Deepfake Detection

CNN-based methods

  • ✔️ MesoNet [paper], [github]
  • Detecting Deep-Fake Videos from Phoneme-Viseme Mismatches, [paper]
  • Deep Fake Image Detection Based on Pairwise Learning, [paper]

RCN-based methods

  • Recurrent Convolutional Strategies for Face Manipulation Detection in Videos, [paper]

Other ML methods

  • SVM: Exposing Deep Fakes Using Inconsistent Head Poses, [paper]

Datasets