The new CompreFace 1.0 version release

Share article

Face Recognition Coronavirus

We have finally released our 1.0 version of Compreface. Try it out and share your feedback with us!

Here is what we added in our 1.0 release, plus other cool news you might have missed from the last release.

Manage Collection 

Now you can manage subjects and images with our user-friendly UI.

The Manage Collection page functionalities include:

  1. Create, delete, rename, and merge subjects
  2. See the subject examples, delete wrong images, and upload new images

We expect this functionality will be useful for many cases:

  1. You can now initialize the facial collection without any scripts or code
  2. If you have any problem with recognition, you can always check to ensure there are correct images saved in each subject
  3. It is always much easier managing the subject from UI rather than using code


Postman documentation

It is always time-consuming to implement all the REST API requests on your programming language, or just to test it using REST API tools. This is why we released Postman documentation.

With Postman documentation, you can test CompreFace REST endpoints right away with predefined methods. But even more exciting, it automatically generates code snippets in your favorite language, so you don’t need to write this code yourself.

Help with AWS

From the beginning, we developed CompreFace to be cloud and Kubernetes ready. Still, every project required preparations to deploy it. To speed up project development, we created deployment instructions for AWS and CloudFormation script, so it will be easier to install and use CompreFace in Amazon AWS.

Single docker image

CompreFace was created using best practices of modern software development. This means that we split it into microservices and made it Kubernetes- and cloud-ready, and the default version is delivered as docker-compose configuration. However, this configuration was a problem in some cases, so we built a single docker image version of CompreFace.

If you want to read more about single docker image installation and other installation options, please read this guide.


CompreFace is a general tool for facial recognition, which means that you can use it for dozens of business cases. But you still need a custom logic for each of these cases.

Fortunately, our community has already created integrations that allow you to use facial recognition with CCTV without writing any custom code.

To use CompreFace for recognition from CCTV, you can use the open-source library Double Take. It connects Frigate, an NVR with real-time local object detection, with CompreFace for facial recognition.

If you use a Home Assistant, open-source software for home automation, you can integrate it with CompreFace and Double Take addons. In this case, you will have a powerful facial recognition from CCTV in one place, managed by Home Assistant.

If you use an unRAID, a file server system, you can integrate it using CompreFace and Double Take templates.

We want to thank all the contributors that help to develop and improve open-source.

Join us! Exadel provides cutting-edge products and services. Check out our open positions here.

Visit this page to find out more about CompreFace

Don’t miss the latest articles and news about CompreFace