Inventor(s)

Qiujun Zhao

Abstract

Customers of online communication and collaboration services (which bring together different capabilities such as video conferencing, online meetings, screen sharing, webinars, Web conferencing, and calling) have been asking for more control over the appearance and the layout of their meetings. In particular, those customers wish to create a custom layout governing how video and shared content will be displayed in a meeting window. Current collaboration services lack such a capability. To address that lack, techniques are presented herein that support the generation of a stage view. Importantly, the generation of such a view requires minimal or no user interaction. Aspects of the presented techniques may allow a user to upload an image to customize a stage background, may employ elements of deep learning to automatically detect and recognize all of the different objects that are depicted in an uploaded image, may offer predefined rules to classify the type of the recognized objects as either background or foreground, may employ image matting algorithms in support of such a background-foreground segmentation, and may employ predefined rules to set the priority of the different background objects to automatically display meeting content.

Creative Commons License

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

Share

COinS