The CompreFace 1.1. Release: What’s New?

We’re excited to announce the release of CompreFace 1.1. In this release, we enhanced user experience, added a new Pose plug-in, introduced a service dashboard, and much more!

New design

This is our second huge redesign of CompreFace. This time, we concentrated on improving your experience. We interviewed users, asked our UI/UX experts, and spent weeks brainstorming how to make CompreFace more user-friendly. We hope you like it, and we’re looking forward to hearing from you how we can further improve going forward!

Pose plug-in

The pose plug-in allows you to get the face’s orientation in 3D space. The roll, yaw, and pitch angles in degrees describe the response result. This plug-in is useful for:

1. Determining if the person is looking straight at the camera. This increases the accuracy of facial recognition.
2. Creating an active liveness detection solution. You can check if the person is real by asking them to turn their head in a certain way, then using the pose plug-in to see if they really did so.
3. Creating a hands-free control system. You can use the pose plug-in to determine how a person is moving their head and create a system that uses that motion as commands.
4. Determining what someone is interested in. You can use the pose plug-in to see what product or advertisment a person is looking at.

The plug-in is available via REST API. We will add its support to our UI in one of the next releases.

Wizard for creating your first facial recognition service

As CompreFace functionality becomes more sophisticated, we want people’s first experience using it to be as straightforward as possible. This is why we created a wizard that will help you decide what type of service to choose and help you create your first facial recognition service.

Facial Recognition Service Dashboard

It’s useful to have all the information about the service in one place. This is why we introducing the Facial Recognition Service Dashboard. You can find tons of helpful information there, like API Key, number of subjects, and number of saved subject examples. You can also use the dashboard to monitor service usage in production.

Registration and session improvements

We simplified the process of first user registration. Now when a user registers, they don’t need to log in; CompreFace automatically redirects the user to the wizard or to the list of applications. If a user is still working with CompreFace and doesn’t close the browser tab, CompreFace won’t invalidate the session.

More custom configurations

We added three more parameters to the .env file, so you can easily customize CompreFace for your needs without rebuilding it:

1. max_file_size and max_request_size options allow you to set the max image size that can be sent to CompreFace
2. uwsgi_processes and uwsgi_threads options let you set the concurrency of the core server that is responsible for running neural networks
3. connection_timeout and read_timeout options allow you to set the max time to wait for the response

See more information about those options and other configuration options here.

We’re excited about these changes and hope that you are, too!

Visit this page to find out more about CompreFace.

Don’t miss the latest articles and news about CompreFace.