Naked-eye 3D Display Viewpoint Conversion via Depth Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current naked-eye 3D display technologies require extra TV resources for viewpoint conversion, leading to reduced image definition and inefficient use of resources.
Innovation Solution
A method and system that decodes 3D signals, separates and improves images, estimates depth information, and interpolates viewpoints to create a composite image, which is then calibrated and transmitted for high-definition naked-eye 3D display, utilizing signal processing techniques like luminance partition scaling and hierarchical interpolation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If viewpoint conversion is performed using conventional methods, then multiple viewpoints can be generated for naked-eye 3D display, but extra TV resources are consumed and image definition is reduced
Solution Approach 1:
The patent segments the image processing into distinct modules: depth extraction module, viewpoint estimation module, and image synthesis module. Each module handles specific tasks independently, allowing efficient resource utilization while maintaining high image definition through specialized processing for each function.
Solution Approach 2:
The patent changes processing parameters by using depth information as a key parameter to control viewpoint synthesis. By adjusting depth maps and using them to guide pixel displacement and blending operations, the system achieves high-definition multi-viewpoint generation without excessive resource consumption.
2Adaptability or versatility
If viewpoint conversion is performed using conventional methods, then multiple viewpoints can be generated for naked-eye 3D display, but TV resources are used inefficiently
Solution Approach 1:
The patent performs preliminary depth extraction and analysis before viewpoint synthesis. By pre-processing the image to obtain accurate depth maps and identifying key objects, the system prepares data structures that facilitate efficient viewpoint conversion, reducing computational overhead during the actual synthesis process.
Solution Approach 2:
The patent uses depth information to create virtual copies of objects at different positions for multiple viewpoints. Instead of processing entire images for each viewpoint, it selectively copies and relocates object pixels based on depth data, significantly reducing computational resources while maintaining image quality.
Data Source
AI summary
The present invention provides a naked-eye 3D display method and system thereof, decoding a 3D signal and obtaining an original 3D video signal; through executing signal separation, image improvement, classifying and frame interpolation to the original 3D video signal, before obtaining the according L image and R image; obtaining the depth information of the objects in the L image and R image and estimating the depth information of each pixel, and forming a corresponding figure of depth; estimating an image with a plurality of L viewpoints according to the L image and the corresponding figure of depth, and estimating an image with a plurality of R viewpoints according to the R image and the corresponding figure of depth; forming a composite image including a plurality of viewpoints after interweaving the images with a plurality of L viewpoints and the images with a plurality of R viewpoints; executing frame interpolation, frequency doubling and image calibration to the composite images, before transmitting to the naked-eye 3D displayer, to achieve the naked-eye 3D display; it has achieved the viewpoints conversion for naked-eye 3D effectively, taking up less resources, having a high definition 3D display, which is stable and fluency, it has a low cost and a wide application.