The July 20, 2022 episode of WebRTC Live was the the third installment of our WebRTC.ventures Roundtable series. Arin welcomed two more members of our expert team to share a nugget of WebRTC wisdom. WebRTC Developer Hamza Nasir spoke about “Attention Detection in Video Conferencing.” Senior DevOps Engineer Hector Zelaya explored “Automating Configuration for WebRTC.”
Attention Detection in Video Conferencing
Covid has changed the paradigm of how the world functions. More and more environments use video conferencing, from schools to business meetings to large conferences. In many of these settings, it is both important and quite useful to be able to gauge participants’ attention to the matter at hand.
In his presentation, Hamza explores how we can calculate a person’s attention using Head-Pose estimation. He uses TensorFlow’s face landmark detection library to leverage its deep neural network to collect data points on a human face and use that to estimate the head’s pose in a given frame. We can then use the Yaw, Pitch and Roll of the face to calculate the attention metrics for a given participant.
Hamza demos this functionality in a video conferencing solution built with the Vonage Video APIs.
Questions revolved around privacy concerns, cultural differences in head poses, the effect of multiple monitors, and whether face masks would skew the results.
For reference:
- Attention Span Prediction Using Head-Pose Estimation With Deep Neural Networks (IEEE paper)
- Hamza’s Demo on GitHub
Automating Configuration for WebRTC
WebRTC enables real-time communication between peers. However, there is a lot of complexity behind such functionality which will ultimately depend on the nature of each specific project. Part of that complexity comes from provisioning and configuring the infrastructure that will support it.
When thinking about running a WebRTC solution at scale, automation can be helpful to reduce the time and efforts on tasks related to such provisioning and configuration, which in turn will allow more focus on the features that will set the solution apart from the others.
In his presentation (see it here on GitHub), Hector provides a high-level overview of his favorite tools (Terraform, Ansible) and techniques that can be used to automate configuration for WebRTC solutions. He also gives us a demo based on a Janus CoTurn server with AWS. It can be used with any open source media server or CPaaS.
After the demo, Arin commented how useful this configuration work is for our WebRTC development services at WebRTC.ventures. It helps us get clients to market faster. It can be used for scaling, testing, and much more. Audience member Dan Jenkins commented that you can spend 2000% of the time you normally would setting up the automation, but now it works every time! It is both reliable and repeatable.
Watch Episode #69!
UP NEXT! WebRTC Live #70 with Kamailio Consultant, VoIP Engineer, and SIP Expert Fred Posner
Wednesday, August 24 at 12:30pm Eastern.
Register today!
Do you have a topic that you would like to see discussed on WebRTC Live? Let us know by emailing news@webrtc.ventures.
Never miss an episode of WebRTC Live, our webinar series hosted by WebRTC.ventures Founder and CEO, Arin Sime. We feature the latest use cases and technical updates to this increasingly popular coding standard for live video. Watch past episodes on our WebRTC Live page, our YouTube channel, and on our blog. Better yet, use the form in the sidebar to join our mailing list and be among the first to hear about upcoming episodes and the latest news in WebRTC!
