Stereo Image Depth Map Error Correction via Spatial Histograms

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current technologies for converting 2D image content to 3D image content often result in errors due to inaccuracies in depth map information, which can diminish the 3D effect experienced by viewers.

Innovation Solution

A method and apparatus that detect and correct errors in depth map information by analyzing spatial histograms and peak frequencies, allowing for the accurate conversion of 2D image data into 3D image data, thereby enhancing the perceived 3D effect.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If depth map information is used to convert 2D image content to 3D image content, then 3D image content can be provided, but errors occur in the depth map information that reduce the 3D effect

Engineering Contradiction:
Improve3D image content provisionVSAvoiddepth map information accuracy
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent applies feedback by comparing the depth map information with the 2D image content and using this comparison to detect and correct errors. The system generates corrected depth map information based on the comparison results, creating a feedback loop that improves accuracy while maintaining 3D content provision capability

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces an intermediary correction process that mediates between the original depth map information and the 2D image content. This intermediary step involves detecting errors through comparison and generating corrected depth map information, which then serves as the basis for accurate 3D image content provision

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of manufacture

If depth map information is applied to 2D image content to produce 3D image content, then 3D effect is achieved, but errors in depth map information diminish the 3D effect

Engineering Contradiction:
Improve3D image content generationVSAvoiddepth perception accuracy
Core Design Contradiction:
Ease of manufactureVSManufacturing precision

Solution Approach 1:

The patent applies preliminary action by performing error detection and correction on the depth map information before it is applied to the 2D image content. The system compares depth map information with 2D image content in advance, detects errors, and generates corrected depth map information that ensures accurate depth perception in the final 3D output

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from comparing depth map information with 2D image content to identify and correct errors. This feedback mechanism ensures that the depth map information used for 3D image generation is accurate, thereby maintaining manufacturing precision while preserving ease of content generation

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2560398B1Method and apparatus for correcting errors in stereo images
Publication Date: 2019.08.14 SK PLANET CO LTD
  • EP2560398B1 patent drawingFigure 1
  • EP2560398B1 patent drawingFigure 2
  • EP2560398B1 patent drawingFigure 3(A)~4

AI summary

An embodiment of the present invention relates to a method and apparatus for correcting errors in stereo images. The apparatus for correcting errors in stereo images according to an embodiment of the present invention comprises: a space histogram generation unit generating space histogram information using the depth map information on the input image data; a peak frequency generation unit generating a peak frequency using the 2D image data of the input image data; an object analysis unit determining the error in each frame of the input image data on the basis of the space histogram and peak frequency; a depth map error correction unit correcting the depth map information to reduce the error; and a rendering processing unit generating left and right eye images, which are stereo images, by using the corrected depth map information. According to an embodiment of the present invention, errors in depth map information can be determined and reduced so as not to reduce stereoscopic perception due to the errors in the depth map information from the process of providing stereo image contents in which the depth map information is applied to the input 2D image contents.