pillbox
faceswap-GAN
pillbox | faceswap-GAN | |
---|---|---|
2 | 1 | |
21 | 3,328 | |
- | - | |
0.0 | 0.0 | |
about 2 years ago | about 2 years ago | |
Jupyter Notebook | Jupyter Notebook | |
- | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
pillbox
faceswap-GAN
-
[D] How is it checked if models do not just memorize their training examples?
But there's a nice survey on Arxiv here of various deepfake / face swap methods. Some of methods listed in the table on page 4, such as Faceswap and Faceswap-GAN, apparently use encoder-decoder models. I think Faceswap-GAN was the one that I was thinking of in particular; apparently it adds a perceptual loss and an adversarial loss to an autoencoder.
What are some alternatives?
AvatarGAN - Generate Cartoon Images using Generative Adversarial Network
faceswap - Deepfakes Software For All
daydreamer - DayDreamer: World Models for Physical Robot Learning
CycleGAN - Software that can generate photos from paintings, turn horses into zebras, perform style transfer, and more.
nn - 🧑🏫 60 Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠
pytorch-CycleGAN-and-pix2pix - Image-to-Image Translation in PyTorch
DCT-Net - Official implementation of "DCT-Net: Domain-Calibrated Translation for Portrait Stylization", SIGGRAPH 2022 (TOG); Multi-style cartoonization
ydata-synthetic - Synthetic data generators for tabular and time-series data