Extended Scene Viewing With Selective Stream Stitching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for creating large fields of view from individual camera views, such as 360° surround views, require significant bandwidth for real-time transmission, limiting their viewing contexts and equipment.
Innovation Solution
A method and device that receive individual video streams at a content consumption device, combine them to fit a display format, and navigate within the scene using user input, reducing bandwidth requirements by only receiving necessary streams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If stitched scenes are transmitted in real time with large high resolution frames, then the field of view coverage is improved, but the bandwidth requirement increases significantly
Solution Approach 1:
The patent divides the large high-resolution stitched scene into multiple smaller lower-resolution video streams, each covering a specific portion of the overall scene. Instead of transmitting the complete large scene, the system segments it into manageable parts that can be selectively transmitted and combined at the receiving end, thereby reducing bandwidth requirements while maintaining the capability to display large field of view content.
Solution Approach 2:
The patent implements dynamic selection and combination of video streams based on the current view requirements. The system dynamically determines which portion of the stitched scene is currently needed and selectively receives and combines only the relevant video streams, rather than continuously transmitting and processing all streams. This dynamic approach reduces bandwidth consumption while maintaining real-time viewing capability.
2Loss of information
If all video streams are received and combined to display the complete scene, then the scene completeness is improved, but the processing complexity and bandwidth usage increase
Solution Approach 1:
The patent extracts and processes only the necessary portion of the stitched scene that corresponds to the current view requirements. Instead of receiving and processing all video streams to maintain complete scene data, the system extracts only the relevant streams needed for the current display, reducing processing complexity and bandwidth usage while maintaining the ability to display complete scenes when needed.
Solution Approach 2:
The patent applies partial action by receiving and processing only a subset of the available video streams at any given time, rather than processing all streams continuously. The system determines the minimum necessary streams required to construct the current view and processes only those, reducing computational complexity and bandwidth consumption while maintaining scene completeness for the current view.
3Adaptability or versatility
If the complete stitched scene is transmitted, then the viewing flexibility is improved, but the bandwidth and equipment requirements increase
Solution Approach 1:
The patent enables viewing flexibility through dynamic stream selection and combination. The system can adaptively receive, process, and combine different combinations of video streams based on the current viewing requirements, allowing users to navigate and view different portions of the stitched scene as needed. This dynamic approach maintains viewing flexibility while reducing bandwidth requirements by transmitting only the necessary streams for the current view.
Data Source
AI summary
The disclosure provides methods and content consumption devices that enable a scene, for example a 360° scene, that is larger (i.e. has more pixels in at least one dimension) than a display format of the content consumption device to be displayed. Constituent scene views are received individually by the content consumption device, for example as broadcasts, and are combined, for example stitched together, at the content consumption device to output a part of the scene that fits in the display format. The part of the scene (and hence the required constituent streams) to be displayed are determined by a signal, for example a navigational input from a user, enabling the user to navigate in the scene. By, for example, receiving a limited number of scene streams based on the scene streams required to produce the view that fits in the display format based on the signal, bandwidth requirements to view and navigate within the scene.