AFHQv-2

AFHQv-2 is a dataset that contains high-quality images of animals, specifically dogs, cats, and wild animals, and is used to evaluate the performance of generative models in image synthesis tasks.

Papers using AFHQv-2 (5)

Variational Trajectory Optimization of Anisotropic Diffusion Schedules2026RODS: Robust Optimization Inspired Diffusion Sampling for Detecting and Reducing Hallucination in Generative Models2025Learning to Discretize Denoising Diffusion ODEs2024Generative Multiplane Images: Making a 2D GAN 3D-Aware2022UDPM: Upsampling Diffusion Probabilistic Models2023