vision-aided-gan VS projected-gan

If you are looking for more traditional noise -> xxx GANs, go for https://github.com/autonomousvision/projected_gan/. Another recent work is https://github.com/nupurkmr9/vision-aided-gan.

Interactive GAN NFT trained with a vision-guided StyleGAN2 using this repo. Every nug/strain generation is unique and you can also create interpolations (takes around 2 minutes to render) between two nugs/strains. If you would like to help me distribute these please specify whether you would like 50, 100, 250, or 500! Thank you so much!

Github: https://github.com/nupurkmr9/vision-aided-gan

Use the images as a training dataset for this

If you are looking for more traditional noise -> xxx GANs, go for https://github.com/autonomousvision/projected_gan/. Another recent work is https://github.com/nupurkmr9/vision-aided-gan.

Made with two ml algorithms, one that generates a tiny 256x256 'base' image, and another that upscales it. Both of them have been trained on a astrophotography-based dataset, but the model architecture are open sourced from Projected-Gan and Real-ESRGAN respectively.

github: https://github.com/autonomousvision/projected_gan

Projected GANs are maybe the currently best model for small datasets. You could also check out StyleGAN 2 ada. https://github.com/autonomousvision/projected_gan

Researchers from the University of Tübingen, Max Planck Institute for Intelligent Systems, and Heidelberg have studied ways to improve GAN training by using pre-trained representations. The researchers proposed a more effective strategy (Projected-GAN) that combines features across channels and resolutions.