Stefan LindmarkFollow


A technique is proposed for optimizing the bandwidth usage of a video conferencing system. The video quality requested and received by a client device for specific video streams associated with a client device user-interface (UI) can be lowered based on user engagement, thereby lowering total bandwidth usage. An eye-tracking module can be used to identify the specific area of visual focus of a user. A ranking module can then generate rankings for every visual element (and any associated video streams) of the UI. Video quality adjustments can then be made based on such rankings. A quality adjustment could include lowering video resolution, lowering video frame rate, lowering compression quality, or any combination of such. Thus, overall bandwidth usage by a client device, and a corresponding conferencing platform, can be reduced.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.