FFHQ-256

The 'FFHQ-256' dataset contains high-quality images of human faces and is used to evaluate the performance of image generation models, particularly in terms of image quality and convergence speed.

Papers using FFHQ-256 (2)

Compensation Sampling for Improved Convergence in Diffusion Models2023MCGAN: Enhancing GAN Training with Regression-Based Generator Loss2024