Real-time face recognition systems are still a very hot topic in computer vision, and many companies have developed their own solutions to try to tap into the growing market. Compared with traditional recognition methods, real-time face recognition system has the advantage of using multiple instances of the same person in consecutive frames.

If you want to take advantage of real-time face recognition, an open source project might be a good place to start. Since the source code has been released, you can see how it works and be sure it doesn’t steal your data. In this article, we’ll help you pick the best open source face recognition projects and show you why choosing open source software is often the best choice.

Why open source face recognition projects?

Open source software has many advantages. First, with open source code, you can determine how the data will be processed. Second, open source projects tend to be of higher quality. Because multiple developers are constantly reviewing the code, errors can be identified very quickly. Third, with low licensing fees, such projects are often developed in-house or by a discretionary IT service provider.

It’s hard to find outdated open source software because it usually follows modern software development practices. Finally, open source is considered the next level of code maturity. It allows developers to understand the code fluently in a matter of minutes and motivates them to work.

Best open source face recognition software

We researched the Github repository of real-time open source facial recognition software and prepared a list of the best choices:

1.Deepface

The library supports different facial recognition methods, such as FaceNet and InsightFace. It also provides a REST API, but it only supports validation methods, so you can’t create a collection of faces and find faces in it. While It’s easy for Python developers to get started, others may have a harder time integrating. The latest version, as of early 2021, is 0.0.49.

2.CompreFace

The solution was only released on Github in July 2020, and it looks very promising. CompreFace makes our list of the best open source face recognition projects because it is one of the few self-hosted REST API face recognition solutions that can be launched with a Docker-compose command.

REST apis allow you to easily integrate it into your systems without prior machine learning skills. Plus, it’s scalable, so you can recognize faces in multiple video streams at the same time.

CompreFace has a simple UI for managing user personas and facial sets. It offers a choice between the two most popular facial recognition methods: FaceNet (LFW 99.65% accuracy) and InsightFace (LFW 99.86% accuracy). It is still in active development, with the latest release as of early 2021 being version 0.5.

3.Face Recognition

The main feature of this solution is its use of their Python API and binary command line tools. In addition, their Github provides installation instructions for all major platforms, even a Docker image for quick setup.

Despite its popularity, the software has some disadvantages. The last release was in 2018 and there have been no major improvements since. It uses a rather outdated face recognition model, is only 99.38% accurate on LFW, and has no REST API.

4.InsightFace

InsightFace is another open source Python library that uses one of the latest and most accurate face recognition methods for face detection (RetinaFace) and face recognition (subcenter-ArcFace). This solution has a very high accuracy rate — 99.86% on the LFW dataset. The only downside is that it’s not easy to use.

5. FaceNet

FaceNet is a popular open source Python library. The accuracy of this method is quite high — 99.65% on the LFW dataset, which is good but not the highest. The downside of this solution is that it has no REST API and no longer supports repositories (last updated in April 2018).

6.InsightFace-REST

This is another promising repository created in 2019, with active development beginning in October 2020. Like CompreFace, this is a Docker-based solution that provides a convenient REST API. The biggest advantage is that its developers have tripled the speed of InsightFace recognition.

The downside of this solution is that it only provides the embedding of faces, there is no API for actual face recognition, so you need to have your own classifier. The repository is still not licensed, so you need to ask the author if it is available. The latest version as of early 2021 is V0.5.9.6.

conclusion

While the best open source facial recognition projects available on GitHub today vary in functionality, they all have the potential to make your development easier.

When choosing an open source face recognition solution, we recommend compiling a list of criteria relevant to your business and choosing options that prioritize the same things you do. While some features may be more important to you than others, each of the free and open source projects we’ve identified here will provide a high-quality real-time face recognition experience.

Combining years of technological experience in the video field, TSINGSEE Video achieves massive video access, intelligent analysis and processing capabilities by using the deployed Al algorithm to process a large number of smart camera video feeds in real time. At present, EasyCVR has achieved face detection, traffic statistics, vehicle detection, license plate recognition and other AI intelligent recognition technology research and development, and is widely used in traffic, logistics, security, fire and other scenarios. In the future, TSINGSEE Video will provide more industry solutions based on deep learning video analysis.