OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Faceaisharp

Faceaisharp is an open-source C# and .NET library for image-based face computer vision. It provides face detection, facial landmark extraction, face-recognition embeddings and eye-state detection, using pretrained ONNX models and ONNX Runtime for inference.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Faceaisharp — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Faceaisharp when building a .NET image-processing feature that needs one or more of its documented face detection, landmark, embedding or eye-state functions.

Who it is for

  • .NET developers adding face-related computer vision to applications
  • Developers building image workflows that need facial landmarks, recognition embeddings or eye-state classification

Documented uses

  • Find face boundaries in images.
  • Extract landmarks such as eye centres, nose tips and mouth corners.
  • Generate and compare face embeddings.
  • Determine whether eyes are open or closed.
Before you choose

Requirements & limits

  • FaceAiSharp.Bundle includes a selection of models; the separate FaceAiSharp package is the documented choice for selecting models or bringing your own.
  • Inference uses ONNX Runtime, with hardware acceleration where possible.
  • Image handling in the documented setup uses ImageSharp.
  • The library is MIT-licensed, while pretrained ONNX models have separate licenses.
  • The README lists different ArcFace model sizes and notes that hardware can affect relative inference efficiency.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the documented Faceaisharp console example

Create a console project with the packages used in the README’s face detection and embedding example.

Before you start

  • .NET command-line tooling
  • Microsoft.ML.OnnxRuntime and FaceAiSharp.Bundle NuGet packages
  • An image for processing; the README example downloads a sample image

Steps

  1. Create a console app with dotnet new console.

  2. Add the packages with dotnet add package Microsoft.ML.OnnxRuntime and dotnet add package FaceAiSharp.Bundle.

  3. Use the README example code to load an ImageSharp image, detect faces, align face images using landmarks, and generate embeddings.

  4. Run the console program with dotnet run.

Expected outcome

The documented example prints face detections and compares two generated embeddings using a dot product.

Guide limitations

  • The README’s comparison thresholds are example code and are not presented as universal identity-matching criteria.
  • This guide follows the published example and does not establish results on other images, models or hardware.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
georg-jung/FaceAiSharp · Official homepage title · retained in its source language
How we review sourcesSuggest a correction