Obviously having a good quality webcam and microphone and an acoustically treated room would be ideal when it comes to video calls, but honestly, how many of us have the time or the money to get all of that setup? Thankfully NVIDIA is looking into the possibility of maybe using AI to help on that front.
The company has announced NVIDIA Maxine, a cloud AI powered platform that will rely on the use of AI to potentially help improve our video conferencing experience. With Maxine, NVIDIA will use AI to help improve certain things like reducing the bandwidth we use for video calls, where it analyzes the key facial points of a person’s face instead of streaming the entire screen.
This will result in smoother video calls, especially for those who do not have a particularly fast internet connection. It can also upscale the resolution of video calls so that even if you have an older or lower-resolution webcam, the software can be used to digitally enhance it. There are also features like auto framing that tries to keep the user in the center of the frame at all times.
Maxine will also introduce things like automatic translation in the form of captions, so that people can speak in different languages and have their words translated and captioned in real time. One of the benefits of Maxine is that it runs in the cloud, meaning that it should work even if you do not have an NVIDIA GPU.
While some of these features are already available on certain video conferencing apps, the fact that NVIDIA does not have any particular vested interest in any of them means that more apps should be able to take advantage of these features in the future.