Stereoscopic Image Rendering Saliency Depth Remapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current stereoscopic image rendering technologies cause discomfort due to the decoupling between eye accommodation and vergence, leading to fatigue as images are presented at a fixed focal distance despite simulating depth perception.
Innovation Solution
A method and system that generate a saliency map to identify the object of focus, redistribute depth values in the depth map to align with the display screen's depth level, and create virtual stereoscopic image frames, reducing the decoupling between accommodation and vergence by remapping the object's depth perception to match the display screen's focal distance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If stereoscopic image frames are rendered with objects appearing in front of or behind the display screen, then depth perception is enhanced, but eye accommodation and vergence become decoupled causing fatigue and discomfort
Solution Approach 1:
The patent applies local quality by selectively adjusting depth values only for salient objects identified through saliency detection, while preserving the original depth perception for other elements. This targeted approach modifies the depth map locally around salient regions, allowing depth enhancement where needed without causing widespread vergence-accommodation conflict across the entire visual field.
Solution Approach 2:
The patent changes depth parameters by redistributing depth values in the depth map to align salient objects with the display screen's focal plane. This parameter adjustment transforms the depth configuration from one that causes decoupling to one that synchronizes accommodation and vergence distances, thereby reducing eye fatigue while maintaining depth perception.
2Ease of operation
If depth values are redistributed to align with the display screen, then vergence-accommodation conflict is reduced, but the original depth perception of objects may be altered
Solution Approach 1:
The patent applies partial action by redistributing depth values only for salient objects identified through saliency detection, rather than uniformly adjusting all objects in the scene. This selective modification ensures that depth perception is preserved for non-salient elements while comfort is improved for prominent objects that would otherwise cause significant vergence-accommodation conflict.
Solution Approach 2:
The patent performs preliminary saliency detection and depth map modification before rendering the stereoscopic image frames. By pre-identifying salient objects and adjusting their depth values in advance, the system ensures that the final rendered images have optimized depth perception that aligns with the display screen, preventing viewing discomfort before it occurs.
3Ease of manufacture
If saliency detection is performed to identify objects of focus, then depth redistribution can be targeted, but processing complexity increases
Solution Approach 1:
The system performs self-service by automatically detecting salient objects through saliency detection algorithms and autonomously adjusting their depth values without requiring manual intervention. This self-service approach streamlines the workflow by integrating saliency detection and depth map modification into an automated pipeline, reducing the need for complex manual processing while maintaining targeted depth adjustment efficiency.
Data Source
AI summary
In some embodiments, a method of rendering a stereoscopic image based on a two-dimensional image frame and a depth map comprises generating a saliency map of the two-dimensional image frame, determining a region where is located an object of focus from the saliency map, modifying the depth map such that a range of depth values in the depth map that is associated the object of focus is redistributed toward a depth level of a display screen, and generating a virtual stereoscopic image frame based on the modified depth map and the two-dimensional image frame. In other embodiments, systems of rendering stereoscopic images are also described.