switchablenorms/CelebAMask-HQ resource

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

2320 stars · 357 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

CelebAMask-HQ is a large-scale face image dataset containing 30,000 high-resolution face images with manually annotated 512x512 segmentation masks across 19 facial attribute classes. It is designed for training and evaluating algorithms in face parsing, face recognition, and GAN-based face generation and editing.

Use cases

• train face parsing models on high-resolution annotated faces
• train GANs for face generation and editing
• evaluate face recognition algorithms
• segment facial components and accessories in images
• interactive facial image manipulation research
• face hallucination model training

When to choose

• you need manually annotated facial segmentation masks at 512x512 resolution
• you are researching face parsing, recognition, or GAN-based face editing
• you need a large-scale face dataset with 19 attribute classes for non-commercial research

When to avoid

• you need a dataset for commercial purposes (license is non-commercial only)
• you need identity or attribute labels (must be requested separately from the CelebA team)
• you need images at resolutions other than 512x512 masks or need video data

Facets

dataset · maturity stable

computer-vision image-processing machine-learning deep-learning computer-vision image-processing machine-learning deep-learning python cross-platform face-parsing face-recognition face-generation face-editing face-segmentation generative-adversarial-network celeba facial-attributes image-dataset segmentation-masks high-resolution-faces gan-training research-dataset non-commercial

Member repositories

For agents

markdown · JSON · MCP: product_card(name="switchablenorms/CelebAMask-HQ")