CelebA-HQ: High-Quality Face Images for Generative Model Training

A dataset of high-quality face images, likely derived from the CelebA dataset. The dataset is associated with the 2017 research paper 'Progressive Growing of GANs for Improved Quality, Stability, and Variation' by Karras et al. It was uploaded to Hugging Face by user Chris1 and last updated on November 18, 2022.

Use Cases

• Training generative adversarial networks (GANs) based on the associated research paper.
• Benchmarking image synthesis quality and stability using high-resolution face images.
• Developing face attribute manipulation models based on the CelebA lineage.
• Studying progressive training techniques for neural networks.

Strengths

• Associated with a seminal 2017 research paper on progressive GAN training.
• Last update timestamp (2022-11-18) is provided.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license information are unknown.

Provenance

Source

Likely derived from the original CelebA dataset, as cited in the associated research paper.

Collection Method

Processed and uploaded to Hugging Face by user Chris1.

Freshness

Last updated 2022-11-18 14:30:59; freshness should be verified.