2D-to-3D Image Conversion with Trajectory-Based Depth Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing 2D-3D conversion algorithms use normalized relative depth prediction, which can result in less realistic into-screen/out-of-screen effects in 3D displays.

Innovation Solution

An image processing method that predicts depth information of moving objects, corrects the depth prediction based on the object's trajectory, and converts 2D images into 3D images with enhanced realism.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If normalized relative depth prediction is used for 2D-3D conversion, then the conversion process can be simplified, but the realism of into-screen/out-of-screen effects deteriorates

Engineering Contradiction:
Improveconversion process complexityVSAvoiddepth prediction accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent segments the depth prediction process into two distinct stages: first obtaining normalized relative depth prediction results, then performing trajectory-based correction. This segmentation allows the system to maintain the simplicity of relative depth prediction while adding a targeted correction mechanism that improves realism without completely redesigning the conversion process.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements a feedback mechanism where the trajectory information of moving objects is used to correct the initial depth prediction results. By continuously comparing predicted depth with actual trajectory data and adjusting accordingly, the system improves depth prediction accuracy while maintaining the overall simplicity of the conversion process.

Inventive Principle:
Principle #23Feedback

2Productivity

If monocular absolute depth prediction is used, then the depth information can be obtained directly, but the prediction accuracy is insufficient for realistic 3D effects

Engineering Contradiction:
Improvedepth information acquisition efficiencyVSAvoiddepth prediction accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary normalized relative depth prediction to obtain initial depth information quickly, then uses trajectory-based correction as a refinement step. This preliminary action approach allows the system to acquire depth information efficiently first, then improve accuracy selectively where needed for realistic 3D effects.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the depth parameter representation from absolute monocular depth to normalized relative depth, then further refines it through trajectory-based correction. This parameter transformation allows the system to balance between acquisition efficiency and prediction accuracy by working with relative values that are then adjusted based on motion trajectories.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20250292424A1Image processing method, electronic device, and storage medium
Publication Date: 2025.09.18 LENOVO (BEIJING) LTD
  • US20250292424A1 patent drawing
  • US20250292424A1 patent drawing
  • US20250292424A1 patent drawing

AI summary

An image processing method, an electronic device, and a storage medium are provided in the present disclosure. The method includes obtaining a plurality of frames of first images; predicting depth information of the moving object to obtain first depth prediction information of the moving object in the plurality of frames of first images; obtaining moving trajectory information of the moving object in the plurality of frames of first images; according to the moving trajectory information of the moving object in the plurality of frames of first images, performing a correction processing on the first depth prediction information of the moving object in the plurality of frames of first images to obtain second depth prediction information of the moving object in the plurality of frames of first images; and converting the plurality of frames of first images into a plurality of frames of second images for three-dimensional effect output.