3D Image Generation from 2D Input Eliminating Black Strips

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for generating 3D images from 2D images often result in black strips at the vertical sides of the left and right eye frames due to shifting operations, which can be distracting and detract from the 3D effect.

Innovation Solution

A method that shifts all pixels in a row by a common amount after a predetermined point, with variable mapping before this point to ensure smooth filling and avoid black strips, using a linear relationship to maintain image integrity and prevent 'stretching' noticeable to the viewer.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If a horizontal shift is applied to generate left and right eye frames from a single 2D frame, then 3D depth effect is achieved, but black strips appear at the vertical sides of the frames

Engineering Contradiction:
Improve3D image generation processVSAvoidblack strips at vertical sides
Core Design Contradiction:
Ease of manufactureVSObject-affected harmful factors

Solution Approach 1:

The image processing is divided into three distinct segments: (1) generating left and right eye frames by sub-sampling and shifting, (2) detecting black strips in the shifted frames, and (3) removing the detected black strips. This segmentation allows each step to be optimized independently, effectively eliminating the harmful black strips while preserving the 3D effect.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

An intermediary detection step is introduced between the shifting operation and the final display. The black strip detection process acts as a mediator that identifies problematic regions caused by the shift operation, enabling their subsequent removal without affecting the overall 3D image generation process.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If the shift amount is increased to enhance 3D depth perception, then the 3D effect is improved, but the width of black strips increases

Engineering Contradiction:
Improve3D depth perceptionVSAvoidwidth of black strips
Core Design Contradiction:
Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The system employs feedback through the black strip detection process that monitors the shifted frames for the presence and width of black strips. This feedback mechanism allows the system to adaptively identify and remove black strips of any width, enabling the use of larger shift amounts for enhanced 3D depth perception without being constrained by the increasing black strip problem.

Inventive Principle:
Principle #23Feedback

3Productivity

If sub-sampling is applied to generate left and right eye frames, then processing efficiency is improved, but image resolution is reduced

Engineering Contradiction:
Improveframe processing speedVSAvoidimage resolution
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies sub-sampling (a form of partial action) to reduce the number of pixels that need to be processed, thereby improving processing efficiency. However, it compensates for the resolution loss by selectively removing only the problematic black strip regions rather than processing the entire frame at full resolution, achieving a balance between productivity and precision.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2509328B1Method and apparatus for generating a 3d image from a 2d image
Publication Date: 2022.01.26 VESTEL ELEKTRONIK SANAYI & TICARET ANONIM SIRKETI
  • EP2509328B1 patent drawingFigure 1
  • EP2509328B1 patent drawingFigure 2
  • EP2509328B1 patent drawingFigure 3

AI summary

The invention provides a method of generating a 3D image from an input 2D image by generating at least one shifted 2D image for presentation to a user in combination with the input 2D image or another shifted image derived from the input 2D image, the method comprising: receiving an input 2D image; for a first row of pixels in the input 2D image, from some selected point along the row mapping all pixels a common number of pixel locations to the left or mapping all pixels a common number of pixel locations to the right thereby shifting the image and forming a shifted image; mapping pixels in locations in the input 2D image before the predetermined point, a variable number of pixel locations to the left or right to positions in the shifted image; and repeating for other rows in the input 2D image.