HaiyuWu/Vec2Face

This is the official implementation of "Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors", which is accepted at ICLR2025.

44
/ 100
Emerging

This project helps researchers and developers in facial recognition to quickly generate large, diverse datasets of face images. By taking an identity's unique "feature vector" (a mathematical representation of their face), it can output many variations of that person's face with different poses and attributes. This is invaluable for training and testing facial recognition systems without needing to collect vast amounts of real-world photos.

Use this if you need to create a large-scale, controlled dataset of synthetic face images for training or evaluating facial recognition models.

Not ideal if you require entirely realistic, unconstrained face images from diverse real-world scenarios or are not working on facial recognition tasks.

facial-recognition synthetic-data-generation computer-vision-research dataset-augmentation
No Package No Dependents
Maintenance 10 / 25
Adoption 9 / 25
Maturity 16 / 25
Community 9 / 25

How are scores calculated?

Stars

86

Forks

6

Language

Python

License

MIT

Last pushed

Jan 19, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/HaiyuWu/Vec2Face"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.