Images

CelebA-HQ

Introduced in Progressive Growing of GANs for Improved Quality, Stability, and Variation2017

About this Dataset

The CelebA-HQ dataset is a high-quality version of CelebA that consists of 30,000 images at 1024×1024 resolution.

Source: IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesis

Image Source: https://github.com/tkarras/progressive_growing_of_gans

Source: Progressive Growing of GANs for Improved Quality, Stability, and Variation

Dataset Variants

CelebA-HQ 512x512CelebA-HQ 256x256CelebA-HQ 1024x1024Celeb-HQ 4x upscalingCelebA-HQ 64x64CelebA-HQ 128x128CelebA-HQ

Papers1

Progressive Growing of GANs for Improved Quality, Stability, and Variation

A new training methodology for generative adversarial networks is described, starting from a low resolution, and adding new layers that model increasingly fine details as training progresses, allowing for images of unprecedented quality.

Dataset Loaders

EDIT

📦

tkarras/progressive_growing_of_gans

tf

🎯

tensorflow/datasets

tfjax

Tasks

EDIT

Image Generation Image Super-Resolution Image-to-Image Translation Image Inpainting Density Estimation Multimodal Unsupervised Image-To-Image Translation Blind Face Restoration Unconditional Image Generation

Similar Datasets

MNIST

CelebA

JFT-300M

Statistics

Papers

1

Tasks

18

Introduced

2017

License

CC BY-NC 4

Modalities

Images