InsightFace

InsightFace is a face analysis toolkit for developers and teams building identity verification, access control, or face editing software. The code uses the MIT license. Its Python tools and self-hosted recognition server run inference on your own hardware. It also offers commercial models and API access for face swapping and deepfake detection.

InsightFace Server runs in a Linux x86_64 container with ONNX Runtime on CPU or NVIDIA GPU. A browser interface, REST API, and Python client cover face detection, comparison, enrollment, and person search. Optional RGB liveness checks assess faces before recognition. The server supports exact searches across enrolled identities and persistent RTSP camera monitoring, with SQLite for local storage.

PrivateFrame handles video face redaction on your device through a desktop app, CLI, or Python. It tracks faces, uses reference photos to select who stays visible, and exports blur or mosaic effects. Saved analysis lets you revise the redaction and render again without repeating face detection.

The Python package works with model packages such as buffalo_l and antelopev2 and supports CoreML, CUDA, and CPU execution. InspireFace provides a C/C++ SDK with anti-spoofing and liveness detection for Linux, Android, iOS, macOS, and embedded devices. Face swapping includes InSwapper-128 and a local NVIDIA GPU option for live streaming. Pretrained models are restricted to non-commercial academic research unless you obtain a commercial license; PrivateFrame also requires a commercial model license.