UniFace

A Unified Face Analysis Library for Python

Get Started View on GitHub

Face Detection

RetinaFace, SCRFD, and YOLO detectors with 5-point landmarks.

Face Recognition

AdaFace, ArcFace, EdgeFace, MobileFace, and SphereFace embeddings for identity verification.

Landmarks

Dense facial landmark localization — 106-point (2d106det) and 98 / 68-point (PIPNet) variants.

Attributes

Age, gender, race (FairFace), and emotion detection from faces.

Face Parsing

BiSeNet semantic segmentation with 19 facial component classes.

Gaze Estimation

Real-time gaze direction prediction with MobileGaze models.

Head Pose

3D head orientation (pitch, yaw, roll) estimation with 6D rotation models.

Tracking

Multi-object tracking with BYTETracker for persistent face IDs across video frames.

Anti-Spoofing

Face liveness detection with MiniFASNet to prevent fraud.

Face Quality

eDifFIQA scalar quality score to filter or rank faces before recognition.

Privacy

Face anonymization with 5 blur methods for privacy protection.

Vector Indexing

FAISS-backed embedding store for fast multi-identity face search.

---

Installation

UniFace uses portable model runtimes for consistent inference across macOS, Linux, and Windows. Most core components run through ONNX Runtime, while optional components may use PyTorch where appropriate.

CPU / Apple Silicon

pip install uniface[cpu]

GPU (NVIDIA CUDA)

pip install uniface[gpu]

From Source

git clone https://github.com/yakhyo/uniface.git
cd uniface
pip install -e ".[cpu]"   # or .[gpu] for CUDA

---

Next Steps

Quickstart

Get up and running in 5 minutes with common use cases.

Tutorials

Step-by-step examples for common workflows.

API Reference

Explore individual modules and their APIs.

Guides

Learn about the architecture and design principles.

---

License

UniFace is released under the MIT License.