Composite Video Creation via Dynamic Peer Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video communication techniques fail to optimize the limited video view by equally presenting each location in a group setting, leading to a disadvantageous focus on recent or frequent speakers, which does not allow for a continuous view of all users, including those who have not spoken.
Innovation Solution
The system determines peers in a video communication, analyzes their video streams to identify users and static locations, modifies the image content by changing shapes and resizing users to uniform sizes, and superimposes modified video streams onto static areas or primary peers based on contextual data and thresholds to create a composite video that includes all participants.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If video streams of all peers are equally presented, then all users can be continuously viewed, but the video view becomes cluttered and difficult to manage
Solution Approach 1:
The system segments video streams by identifying primary peers (those with multiple users or speaking users) and secondary peers (single users who are not speaking). This segmentation allows the composite video to organize streams hierarchically, placing primary peer streams as base layers and superimposing secondary peer streams on top, thus maintaining visibility of all users while preserving video view organization.
Solution Approach 2:
The system applies different processing qualities to different video streams based on their importance. Primary peer video streams are processed with full resolution and clarity, while secondary peer video streams are modified with changes in shape, size, and opacity. This local quality differentiation ensures that important users remain prominent while less important users are still visible but do not clutter the view.
2Productivity
If focus is placed on recent or frequent speakers, then active communication is highlighted, but users who have not spoken are excluded from continuous view
Solution Approach 1:
The system dynamically adjusts the composition of video streams based on real-time communication state. Users who are speaking are identified and their video streams are designated as primary peer streams, while non-speaking users become secondary peer streams. This dynamic reconfiguration ensures that active speakers receive appropriate focus while non-speaking users remain continuously visible in the composite video, maintaining both communication efficiency and inclusivity.
3Ease of operation
If modified video streams are superimposed onto primary peers, then a seamless composite video is created, but the complexity of video processing increases
Solution Approach 1:
The system extracts and removes specific elements from video streams to simplify processing. Background elements are removed from secondary peer video streams before superimposition, and only essential user video content is retained. This extraction approach reduces the complexity of compositing operations while maintaining seamless integration in the final composite video.
Solution Approach 2:
The system modifies video stream parameters such as shape, size, and opacity to facilitate easier compositing. Secondary peer video streams are transformed into standardized formats with adjusted parameters that simplify their integration onto primary peer streams. These parameter changes reduce processing complexity while achieving seamless composite video quality.
Data Source
AI summary
Systems, methods, and non-transitory computer readable media are configured to receive a video stream of a first peer involved in a video communication. A video stream of a second peer involved in the video communication is received. Image content of the video stream of the second peer is modified to create a modified video stream of the second peer. The video stream of the first peer and the modified video stream of the second peer are combined to create a composite video.