Composite Video Generation with Dynamic User Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video communication techniques fail to optimize the presentation of multiple users in a group setting, often resulting in a co-equal presentation of each location or highlighting only recent speakers, which does not allow for a continuous view of all users, including those who have not spoken recently or at all.
Innovation Solution
A computer-implemented method and system that analyzes video streams to determine the presence of users and static locations, modifies image content, and composites video streams to create a tailored composite video that accommodates all peers, with features like resizing, shape modification, and superimposition of video streams based on contextual data and affinity analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional video communication techniques present multiple users in a group setting, then each location can be displayed, but the presentation becomes co-equal and fails to provide continuous view of all users with appropriate focus
Solution Approach 1:
The patent applies local quality by assigning different visual priorities to different users in the composite video. Primary users are displayed with full visibility and focus, while secondary users are integrated into the background. This creates differentiated quality levels within the same video presentation, allowing important users to stand out while still maintaining visibility of all participants.
Solution Approach 2:
The system segments users into different categories (primary users and secondary users) based on their importance or activity level. This segmentation allows the video compositing system to apply different processing and display rules to different user groups, enabling continuous viewing of all users while maintaining appropriate focus on key participants.
2Ease of operation
If all users are continuously displayed in the composite video, then user interaction and engagement are enhanced, but background and foreground elements may create distractions
Solution Approach 1:
The patent extracts secondary users from the foreground and places them in the background of the composite video. This extraction technique allows all users to remain visible while removing potential distractions from the primary viewing area. Secondary users are effectively taken out of the main focus zone, reducing visual competition for the viewer's attention.
Solution Approach 2:
By applying different visual treatments to different regions of the video composite, the system enhances user interaction areas while minimizing distracting elements in background regions. Primary users in foreground positions receive full visual quality, while secondary users in background positions are rendered with reduced prominence, eliminating distractions while maintaining continuous visibility.
3Quantity of substance
If video streams are composites with multiple users of different sizes and positions, then all users can be visible, but maintaining approximately the same size and focus for all users becomes difficult
Solution Approach 1:
The system applies local quality by positioning primary users in foreground areas where they receive full visual prominence and appropriate sizing, while secondary users are placed in background areas with adjusted sizing and positioning. This localized differentiation allows all users to be visible with approximately the same effective size and focus, while accommodating the natural variations in user importance and camera perspectives.
Data Source
Figure 1
Figure 2A
Figure 2B
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.