Lenticular Display Distortion Mapping for Crosstalk-Free 3D Viewing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Lenticular distortion in autostereoscopic displays occurs due to manufacturing defects in the lenticular sheet, causing crosstalk between left and right images and resulting in ghost images and eye-strain for the observer.

Innovation Solution

A method and apparatus that utilize a lenticular distortion tracking camera to capture fringe patterns from various angles, generating dual-component lenticular distortion maps based on intensity patterns and lenticular display parameters, which are used to adjust a swizzle function to compensate for distortion, ensuring clear 3D image viewing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If a lenticular display is manufactured uniformly, then the 3D image can be viewed without noticeable distortion, but manufacturing defects in the lenticular sheet cause lenticular distortion and crosstalk between left and right images

Engineering Contradiction:
Improvelenticular sheet manufacturing uniformityVSAvoidimage quality consistency
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The system performs preliminary measurement of the lenticular sheet's actual characteristics by capturing images at multiple angles and generating distortion maps before the display is put into service. This allows the swizzle function to be pre-calibrated to compensate for manufacturing defects, ensuring reliable image quality from the start.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors and measures the actual lenticular distortion by capturing fringe patterns at various viewing angles and uses this feedback to generate updated distortion maps. These maps are fed back into the swizzle function to dynamically adjust and correct distortion in real-time, maintaining image quality consistency despite manufacturing variations.

Inventive Principle:
Principle #23Feedback

2Reliability

If lenticular distortion is present, then crosstalk between left and right images occurs, but correcting distortion requires capturing images from multiple angles and generating complex distortion maps

Engineering Contradiction:
Improveelimination of crosstalk and ghost imagesVSAvoiddistortion correction system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The distortion correction is segmented into independent angular components by capturing images at multiple discrete angles (first angle, second angle, third angle) and generating separate distortion maps for each angle. This segmentation allows the complex 3D distortion field to be broken down into manageable 2D maps that can be processed and applied independently through the swizzle function.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The swizzle function serves as an intermediary between the raw display output and the final corrected image. It takes the original image data and the generated distortion maps as inputs, processes them through mathematical transformations, and outputs the corrected image that compensates for lenticular distortion, thereby eliminating crosstalk and ghost images.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If distortion correction is implemented in real-time, then clear 3D image viewing is enabled, but the system requires continuous image capture and processing

Engineering Contradiction:
Improvereal-time distortion compensationVSAvoidtime for image capture and processing
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system captures fringe patterns and generates distortion maps in advance before the actual 3D viewing occurs. By performing the measurement and correction data generation as a preliminary calibration step, the real-time viewing process does not need to perform these computationally intensive operations, reducing processing time during actual use.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Once the distortion maps are generated during calibration, the correction information is stored and applied continuously throughout the viewing period without requiring repeated image capture and processing. The swizzle function continuously applies the pre-computed correction data, maintaining distortion compensation without the time cost of repeated measurements.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP3922012B1Method and apparatus for correcting lenticular distortion
Publication Date: 2026.02.04 GOOGLE LLC
  • EP3922012B1 patent drawingFigure 1
  • EP3922012B1 patent drawingFigure 2
  • EP3922012B1 patent drawingFigure 3

AI summary

A method includes capturing an image (10a-310d, 410a-410d, and 510a-510d) of a lenticular display (102), generating a lenticular distortion map (205) of the lenticular display using the image, and compensating for a lenticular distortion of the lenticular display using the lenticular distortion map. An apparatus for performing the method includes the lenticular display, a lenticular distortion tracking camera (110), and at least one processor (150).