cmusatyalab/openface

Health v2 · maintenance only

Adoption not part of the score

15438 stars · 3558 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

OpenFace is a free and open source Python and Torch implementation of face recognition based on Google's FaceNet deep neural network. It generates face embeddings from images and supports training classifiers, real-time webcam demos, and LFW accuracy evaluation.

Use cases

• recognize faces in images with deep neural networks
• generate face embeddings from a batch of photos
• train a face classifier to identify people
• run real-time face recognition on a webcam stream
• compare two face images for similarity
• evaluate face recognition accuracy on LFW

When to choose

• you need an open source FaceNet-style face recognition pipeline
• you want to train a custom face identification classifier from your own images
• you need CPU or CUDA inference for face embeddings

When to avoid

• you need actively maintained state-of-the-art face recognition models
• you want a pure modern PyTorch or TensorFlow implementation without legacy Torch dependencies
• your project requires commercial licensing guarantees

Facets

library · maturity maintenance

deep-learning computer-vision image-processing machine-learning computer-vision image-processing deep-learning artificial-intelligence python face-recognition facenet face-embeddings torch lua lfw linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="cmusatyalab/openface")