Stereoscopic Display Relief Mapping for Jerk-Free Rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating multiple views for stereoscopic three-dimensional displays are computationally intensive, leading to jerky frame sequences and increased computing time, especially with high-resolution displays and numerous views.
Innovation Solution
The method employs relief mapping to create a continuous relief card based on depth information, determining color information for only the pixels required by the combination rule, reducing unnecessary computations and optimizing pixel assignment across different viewing positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple views are generated for stereoscopic display, then the quality of three-dimensional impression is improved, but the computing time and processing effort increase significantly
Solution Approach 1:
The patent divides the image generation process into two independent stages: first generating a depth card from the source view, and then using this depth card to generate multiple views. This segmentation allows the computationally intensive depth extraction to be performed only once, while subsequent view generations use the pre-computed depth information to reduce processing time.
Solution Approach 2:
The depth card is generated in advance as a preliminary step before generating the multiple views. This preliminary action stores depth information that can be reused across all view generations, eliminating the need to recompute depth for each view and significantly reducing the computing time for generating N views.
2Measurement precision
If the resolution of the display screen is increased, then the quality of the stereoscopic image is improved, but the computing effort for generating views increases
Solution Approach 1:
The method separates the depth computation from the view generation process. The depth card is computed once at the source view resolution, and then this depth information is used to generate high-resolution views by ray casting through the depth card. This allows high-resolution display without proportionally increasing the computational burden of depth calculation.
Solution Approach 2:
The depth card acts as a reusable template that can be copied and applied to generate multiple views at different resolutions. Instead of computing depth information for each high-resolution view, the method copies the depth card and uses it to guide the generation of high-resolution views, maintaining productivity while improving image quality.
3Measurement precision
If the number of views is increased, then the stereoscopic effect is improved, but the frame sequence regeneration time increases causing jerking
Solution Approach 1:
The depth card is generated in advance and stored, serving as a preliminary computation that enables fast generation of any number of views. When generating N views, the method only needs to perform ray casting operations using the pre-computed depth card, which is much faster than computing depth for each view from scratch, thus maintaining high frame regeneration speed even with many views.
Solution Approach 2:
The patent segments the view generation into a one-time depth card creation followed by rapid view synthesis using ray casting. This segmentation allows the system to handle a large number of views by performing only the lightweight ray casting operation for each additional view, rather than the full view generation pipeline, thereby maintaining high speed for frame sequence regeneration.
Data Source
AI summary
The invention relates to a method for the stereoscopic display of a scene on a grid of picture elements (pixels), based on several views that are combined according to a combination rule to form a combination image. The combination image contains only selected parts or pixels of each view. Furthermore, propagation directions are fixed, which make it possible for a viewer to see in the combination image a different selection of views with his left eye than with his right eye, which results in a visual impression of space.Starting from an original view, in which each pixel contains bits of color information and depth information, the other views are generated, with exclusively such bits of color information being determined that belong to pixels of the views which are displayed in the combination image. The determination of the bits of color information is performed with the aid of a relief card Z(x,y).