Adaptive Parallax Compensation for Multi-View Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for encoding multi-viewpoint video images face efficiency degradation due to errors in camera parameters and encoding distortions, particularly when parallax compensation is constrained on an Epipolar line, leading to increased code lengths for prediction residuals.
Innovation Solution
The approach dynamically adjusts the number of parameters for parallax compensation based on each reference image's characteristics, encoding the parallax-parameter number data and reference image indices to adaptively control the degree of freedom in parallax compensation, allowing for higher encoding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If parallax compensation is constrained on an Epipolar line using conventional methods, then device complexity is reduced, but encoding efficiency deteriorates due to errors in camera parameters and encoding distortions
Solution Approach 1:
The patent applies dynamics by making the parallax compensation method adaptive rather than fixed. The system dynamically selects between Epipolar line constraint and general parallax compensation based on reference image quality assessment. This allows the encoding system to flexibly adjust its complexity level according to actual conditions, improving encoding efficiency while maintaining reasonable device complexity.
Solution Approach 2:
The patent changes the parameter of parallax compensation degree by introducing a selection mechanism. When reference image quality is high, the system uses full parallax compensation (higher degree of freedom). When quality is low, it switches to Epipolar line constraint (lower degree of freedom). This parameter change resolves the contradiction by adapting the compensation level to actual conditions.
2Measurement precision
If the number of parallax parameters is increased to improve accuracy, then prediction accuracy improves, but code length for prediction residuals increases
Solution Approach 1:
The patent directly applies parameter changes by adjusting the number of parallax parameters based on reference image quality. When reference images are of high quality, the system uses more parameters (higher accuracy) because the benefit outweighs the cost. When reference images are of low quality, it uses fewer parameters to avoid wasting code bits on inaccurate predictions. This resolves the contradiction by making parameter count adaptive to actual conditions.
Solution Approach 2:
The patent applies partial action by selectively using full parallax compensation only when reference image quality justifies it. Instead of always using maximum parameters, the system uses partial compensation (Epipolar constraint) when sufficient, and full compensation only when necessary. This optimizes the balance between accuracy and code length.
3Productivity
If adaptive selection of parallax compensation methods is implemented, then encoding efficiency improves, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the reference images into different quality groups (first group with higher quality, second group with lower quality). This segmentation allows the system to apply different parallax compensation strategies to different groups, improving encoding efficiency without requiring complex adaptive logic for all images uniformly. The segmentation simplifies the decision-making process.
Solution Approach 2:
The patent applies local quality by treating different reference image groups with different quality levels differently. High-quality reference images receive full parallax compensation treatment, while lower-quality images receive constrained parallax compensation. This local differentiation improves overall encoding efficiency by optimizing each group according to its specific characteristics rather than applying a uniform approach.
Data Source
AI summary
A video encoding method for encoding video images as a single video image by using parallax compensation which performs prediction by using parallax between the video images, and a corresponding decoding method. The number of parameters as parallax data used for the parallax compensation is selected and set for each reference image. Data of the set number of parameters is encoded, and parallax data in accordance with the number of parameters is encoded. During decoding, parallax-parameter number data, which is included in encoded data and designates the number of parameters as parallax data for each reference image, is decoded, and parallax data in accordance with the number of parameters is decoded, where the parallax data is included in the encoded data.