deepinsight/insightface

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

29939 stars · 6107 forks observed · 2026-10-12

What it is AI-extracted, prompt v1, taxonomy v1, 2026-10-05, confidence 0.97

InsightFace is an open-source Python library for state-of-the-art 2D and 3D face analysis, including detection, recognition, alignment, swapping, and optional liveness detection. It also provides a self‑hosted server and CLI tools for video privacy and large‑scale face search.

Use cases

• detect faces in images or video streams
• recognize and compare faces for authentication or identity verification
• align facial landmarks for preprocessing in computer‑vision pipelines
• swap faces or apply privacy blurring/mosaic in video content
• run liveness checks to prevent spoofing attacks
• search millions of face embeddings on a single GPU
• integrate face analysis into Python applications via a simple API

When to choose

• You need a high‑accuracy, production‑ready face analysis solution with multiple model options.
• You prefer a Python‑centric library that works on CPU, CUDA, or CoreML without writing low‑level code.
• You want an easy‑to‑deploy self‑hosted server or CLI for batch processing and video privacy.

When to avoid

• You require a fully commercial license for the models without contacting the InsightFace team.
• You need a non‑Python implementation or deep integration with other languages.
• Your use case demands real‑time processing on low‑power edge devices without GPU support.

Facets

library · maturity active

computer-vision image-processing nlp cli artificial-intelligence computer-vision data-science python windows arcface retinaface inswapper privateframe face-analysis onnx sqlite server cli linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="deepinsight/insightface")