UniFace

All-in-One Open-Source Face Analysis Library

ONNX-optimized detectors (RetinaFace, SCRFD, YOLO) with 5-point landmarks.

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

Accurate 106-point facial landmark localization for detailed face analysis.

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

BiSeNet semantic segmentation with 19 facial component classes.

Real-time gaze direction prediction with MobileGaze models.

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

Face liveness detection with MiniFASNet to prevent fraud.

Face anonymization with 5 blur methods for privacy protection.

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

Installation

UniFace runs inference primarily via ONNX Runtime; some optional components (e.g., emotion TorchScript, torchvision NMS) require PyTorch.

Standard

pip install uniface

GPU (CUDA)

pip install uniface[gpu]

From Source

git clone https://github.com/yakhyo/uniface.git
cd uniface
pip install -e .

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

Step-by-step examples for common workflows.

Explore individual modules and their APIs.

Learn about the architecture and design principles.

UniFace is released under the MIT License.