Zoom Certified Presence app: EMOJ biometric recognition within Zoom meetings

Emoj - Human Before Artificial

Emoj develops monitoring and recognition systems for human factors through the use of technologies such as Computer Vision and Machine Learning.

Needs

Emoj's needs are to provide its biometric recognition services to organisations that deliver their services through online meetings.

Let's think, for example, of a university that provides online courses and lessons in video calls with mandatory attendance, for which it is necessary to certify the presence of students during the meeting. Or an organisation that holds courses on safety and must ensure the presence of participants in the lessons for the issue of the certificate.

Emoj's goal is to integrate its biometric identity verification services, via facial photography, with one of the leading video telephony services: Zoom.

The solution

Zoom App “Certified Presence” by Claranet

In addition to the video telephony service, Zoom offers the possibility of developing applications through a developer platform. The applications, downloadable from a marketplace once published, are executed within the Zoom client during meetings and through the Zoom SDK tool it is possible to make the app interact with the client itself.

Claranet has developed the Zoom Certified Presence app that integrates with the Emoj biometric recognition REST API, which allows you to perform identity checks on meeting participants.

Certified Presence App Workflow

How does biometric facial verification work during the meeting?

“Before using the Certified Presence app, users will need to register on the Emoj web portal where they can upload photos of their face and an identity document. Users are then enabled to use the Certified Presence app on Zoom during meetings.”

During Zoom meetings, Meeting Hosts can require participants to complete a biometric verification by opening the Certified Presence app and taking a photo of their face.

Biometric control technologies

Emoj uses a facial recognition system based on a Siamese Neural Network to verify the identity of users during meetings. This type of neural network is often used in scenarios where it is necessary to compare two inputs and determine how similar they are to each other. The term "Siamese" comes from the fact that the network uses two identical subnetworks (with the same weights and parameters) to process the two inputs separately.

When registering on the Emoj web portal, users provide facial images that are stored in a database. When a participant takes a photo during a Zoom meeting, their live image is compared to the stored registration photo.

The system then calculates the distance between the feature vectors of these images. A smaller distance indicates a higher probability of identity matching, confirming that the webinar attendee is the same person who registered.

Infrastructure

“The Certified Presence app was developed using the Fastify framework and deployed on AWS Lambda, a choice determined by the use of AWS as a cloud provider.” The application integrates with REST APIs for identity verification, ensuring a secure and reliable process. Cognito was used as the Identity provider for user management, while S3 was used to store photos uploaded by users during registration.

Terraform was used as the IaC tool for the infrastructure, while AWS CodePipeline was used to manage the CI/CD process.

Result

The collaboration between Emoj and Claranet has created an innovative and highly functional solution for biometric recognition within Zoom meetings, responding precisely to the needs of identity certification in virtual contexts. The integration of advanced technologies such as neural networks and AWS cloud infrastructures guarantees a secure, scalable service in line with the needs of companies and institutions.

Find out more about Claranet's Data and AI solutions, and speak to one of our experts today.