foivospar/Arc2Face

[ECCV 2024 Oral 🔥] Arc2Face: A Foundation Model for ID-Consistent Human Faces ------------------------ [ICCVW 2025] ID-Consistent, Precise Expression Generation with Blendshape-Guided Diffusion

47
/ 100
Emerging

Conditions face generation on ArcFace embeddings extracted from input images, enabling tuning-free personalization without per-subject fine-tuning. Built on Stable Diffusion 1.5 with a custom CLIP text encoder and UNet adapter, trained on WebFace42M for superior ID preservation; supports acceleration via LCM-LoRA, ControlNet integration for pose control, and a new Expression Adapter component that uses blendshape guidance for fine-grained facial expression control.

789 stars.

No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 15 / 25

How are scores calculated?

Stars

789

Forks

55

Language

Python

License

MIT

Last pushed

Oct 10, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/foivospar/Arc2Face"

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