podgorskiy/ALAE

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

3511 stars · 544 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Official PyTorch implementation of Adversarial Latent Autoencoders (ALAE/StyleALAE), a CVPR 2020 paper combining autoencoders with GAN training. It can generate 1024x1024 face images comparable to StyleGAN and perform reconstruction and manipulation of real face images.

Use cases

• generate high-resolution face images
• reconstruct real face images from photos
• style mixing and latent manipulation of faces
• reproduce ALAE paper results
• train GAN-based autoencoders on FFHQ or CelebA-HQ
• learn disentangled latent representations

When to choose

• you need StyleGAN-quality face generation plus real-image reconstruction
• you want to reproduce or build on the CVPR 2020 ALAE paper
• you need an autoencoder with GAN-level generative power in PyTorch

When to avoid

• you need a maintained, actively updated codebase
• you need a license for commercial use (no explicit license file)
• your task is not face/generative image modeling

Facets

library · maturity maintenance

deep-learning machine-learning image-processing computer-vision deep-learning computer-vision image-processing machine-learning python gan autoencoder stylegan face-generation paper-implementation pytorch cvpr2020 generative-models gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="podgorskiy/ALAE")