In this release, we worked hard on a redesign of CompreFace, adding a face mask detection plug-in and more! Let’s take a closer look.
CompreFace now looks much nicer! But this isn’t the only UI improvement:
- With our new responsive design, it’s much more comfortable to use CompreFace on small resolutions; you can even use it on mobile devices.
- We made it easier to see results on images with multiple faces. If there is more than one face in the image, we display results when you hover over the face with your mouse.
- We added age, gender, and landmark results to the UI. Age and gender plug-in results are displayed under the subject name result. And to show the landmark plug-in results, click the landmarks button in the top left corner of the page.
Face mask detection plugin
CompreFace now supports the face mask detection plug-in. With this plug-in enabled, you will receive information if the person is wearing a face mask and if it’s worn correctly.
This feature can be extremely helpful if you have to comply with COVID-19 limitations and want to automate the facemask verification process.
To enable the face mask detection plug-in, just add it into the “face_plugins” parameter. For more information, have a look at our documentation.
We added new endpoints for more flexible work with subjects. Before the 0.6 release, you could only work with examples of subjects. But now we’ve introduced endpoints that directly work with subjects.
Moreover, we added endpoints for downloading saved images. Please note that this feature doesn’t work with the configuration of CompreFace that doesn’t save images.
This functionality is useful if you want to manage a face collection of subjects and show saved images.
We added one more SDK that allows even easier integration with CompreFace — Python SDK. With this face recognition Python SDK, you will be able to integrate face recognition into Python projects without ML skills. All you need is to install CompreFace and add SDK to your project using this command:
pip install compreface-sdk.
Find more information in the SDK repository.
CompreFace has always been a Kubernetes-ready application. We decided to publish Kubernetes configs so you don’t need to write them yourself. Now you can run our face recognition application in any Kubernetes cluster. We also added instructions on how to run it in Minikube, in AWS, and with helm. Find more information in the CompreFace Kubernetes repository.
If you need help with CompreFace, Exadel provides paid support and services. Our services include integration, customization, new feature development, training, and more. To reach us, use the Exadel contact form or email CompreFace.