BigGAN is a type of generative adversarial network that was designed for scaling generation to high-resolution, high-fidelity images. It includes a number of incremental changes and innovations. The baseline and incremental changes are:
The innovations are:
Paper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Generation | 28 | 31.46% |
Conditional Image Generation | 12 | 13.48% |
Super-Resolution | 4 | 4.49% |
Image Super-Resolution | 3 | 3.37% |
Image-to-Image Translation | 3 | 3.37% |
Knowledge Distillation | 2 | 2.25% |
Colorization | 2 | 2.25% |
Image Restoration | 2 | 2.25% |
Model Compression | 2 | 2.25% |