Multi-Viewpoint Image Coding Reference Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-viewpoint image coding methods suffer from reduced coding efficiency due to increased prediction errors when imaging conditions vary, especially when underexposure or overexposure occurs, leading to decreased correlation between images.
Innovation Solution
An imaging apparatus that captures images from multiple viewpoints with exposure parameters set for each viewpoint, selects a reference image based on these parameters, and uses inter-viewpoint prediction to code the images efficiently, thereby reducing prediction errors and improving coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If inter-viewpoint prediction is performed using identical viewpoint image or adjacent viewpoint image as reference images, then coding efficiency is improved under normal conditions, but prediction error increases and coding efficiency decreases when imaging conditions differ (underexposure or overexposure)
Solution Approach 1:
The patent changes the selection criteria for reference images from purely spatial (identical viewpoint or adjacent viewpoint) to include exposure parameter matching. By introducing exposure parameter as a selection criterion, the system adapts the reference image selection to match imaging conditions, thereby maintaining prediction accuracy across varying exposure conditions while preserving coding efficiency.
2Device complexity
If a fixed reference image selection method is used, then device complexity is reduced, but adaptability to different imaging conditions deteriorates
Solution Approach 1:
The patent introduces exposure parameter matching as an additional selection criterion without fundamentally changing the reference image selection architecture. The selection process remains relatively simple by adding a parameter comparison step, while significantly improving adaptability to different imaging conditions including underexposure and overexposure scenarios.
Data Source
AI summary
A multi-viewpoint image which has been captured with respect to an object from a plurality of viewpoints is coded by setting a base viewpoint among the plurality of viewpoints. An image captured from a viewpoint that is not the base viewpoint is then coded using a reference image selected based on imaging parameters and a parallax.