Image Generative Models - Pytorch Implementation from Scratch

• Diffusion Models

  • UNet
  • DDPM (Denoising Diffusion Probabilistic Models)
  • DDIM (Denoising Diffusion Implicit Models)
  • VQVAE (Vector Quantized Variational Autoencoder)
  • VQGAN
  • LDM (Latent Diffusion Model)

• Generative Adversarial Networks

  • GAN
  • DCGAN (Deep Convolutional GAN)
  • WGAN-GP (Wasserstein GAN with Gradient Penalty)
  • SN-GAN (Spectral Normalization for Generative Adversarial Networks)
  • CGAN (Conditional GAN)
  • CycleGAN
  • Pix2Pix
  • MSG-ProGAN (Multi-Scale Gradients for Generative Adversarial Networks)
  • ProGAN (Progressively Growing GAN)
  • StyleGAN
  • StyleGAN 2

• Variational Autoencoders

  • VAE
  • Conditional VAE
  • Adversarial Autoencoder
  • VQVAE (Vector Quantized Variational Autoencoder)

• Autoregressive Models

  • PixelCNN