Multi-view Display Controller View Angle Synthesis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multi-view displays face challenges such as the need for multiple cameras, which are cumbersome, expensive, and difficult to calibrate, and result in varying viewer experiences due to the linear increase in bit rate with the number of views, as well as storytelling issues where actors' directions are hard to manage across different viewer positions.
Innovation Solution
A multi-view display controller that determines and adjusts view angles for each viewer based on their position, allowing for a consistent 3D experience across different angles without the need for multiple cameras, by using a positioning device to track viewer positions and outputting media data with specific view angles to ensure each viewer sees the intended perspective.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a multi-view display uses multiple cameras to capture different views, then the viewer can see content from different angles, but the hardware complexity and cost increase significantly
Solution Approach 1:
The patent uses a single camera to capture 2D video and then creates multiple virtual views through software processing. Instead of using multiple physical cameras, the system generates synthetic views by warping and transforming the single-camera footage, thereby achieving multi-view capability without the hardware complexity of multiple cameras
Solution Approach 2:
The patent replaces the mechanical system of multiple physical cameras with a computational approach using software algorithms. The view synthesis is achieved through digital image processing techniques including warping, blending, and perspective transformation, substituting mechanical complexity with computational processing
2Reliability
If a multi-view display uses many cameras to increase the number of views, then the viewer experience improves, but the bit rate increases linearly with the number of views
Solution Approach 1:
The patent generates multiple views from a single source video through software synthesis rather than capturing multiple independent video streams. This approach creates correlated views that share common information, reducing the total bit rate required compared to transmitting multiple independent camera feeds
Solution Approach 2:
The patent combines multiple views into a single video stream for transmission, using view identification data to indicate which views are available. The system merges the information from multiple synthesized views into one compressed stream, allowing the receiver to extract only the needed views rather than transmitting all views at full resolution
3Ease of operation
If a multi-view display uses a fixed camera rig configuration, then the system is easier to operate, but it causes storytelling issues where actors' directions vary for different viewer positions
Solution Approach 1:
The patent implements dynamic view synthesis where the system can generate and switch between different virtual camera positions and perspectives in real-time. This allows the storytelling perspective to be dynamically adjusted to maintain consistency regardless of the physical camera rig configuration or viewer position
Solution Approach 2:
The patent changes the parameters of virtual camera positions, viewing angles, and perspective transformations through software control. By dynamically adjusting these parameters, the system can maintain consistent storytelling perspectives even when using a fixed physical camera setup, as the virtual camera positions can be reconfigured without moving physical equipment
Data Source
Figure 1~2
Figure 3A~3B
Figure 4~5
AI summary
A multi-view display controller (100, 200, 300) determines view angles for each view of a multi-view media content (10) for each viewer (5, 6) watching a multi-view display (2). The view angles determined for a viewer (5, 6) collectively define a viewer cone (20, 25) that displays the views onto the viewer (5, 6). Media data of the multi-view media content (10) is output together with information of the determined view angles to the multi-view display (2) in order to allow each viewer (5, 6) to have the same experience of displayed media content regardless of where the viewer (5, 6) is positioned relative to the multi-view display (2).