Adaptive Depth Data Filtering for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video signal coding techniques fail to efficiently address the unique characteristics of depth data, leading to suboptimal image quality and coding efficiency when compared to texture data.
Innovation Solution
The implementation of a region-based adaptive loop filter for depth data, with variable kernel sizes and pixel application based on block variance and coding mode, allows for separate and optimized filtering of depth data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a standard in-loop filter is applied to depth data, then block distortion is reduced, but filtering complexity increases and coding efficiency deteriorates
Solution Approach 1:
The patent applies different filtering operations to different regions of the depth data based on block variance. Low-variance blocks receive stronger filtering to reduce block distortion, while high-variance blocks receive minimal or no filtering to preserve details and avoid complexity
Solution Approach 2:
The patent dynamically changes filtering parameters (filter strength, kernel size) based on the variance of each block. This allows the filter to adapt its complexity to the local characteristics of the depth data, reducing overall filtering complexity while maintaining effectiveness where needed
2Manufacturing precision
If a region-based adaptive loop filter with variable kernel size is applied, then image quality improves, but processing time and complexity increase
Solution Approach 1:
The patent applies the computationally intensive variable kernel size filtering only to specific regions where it provides the most benefit (low-variance blocks with visible block distortion). Other regions use simpler, faster filtering operations, reducing total processing time while maintaining image quality where it matters most
3Productivity
If separate filtering for depth data is implemented, then coding efficiency improves, but device complexity increases
Solution Approach 1:
The patent implements a unified filtering framework that handles both texture and depth data, but applies different filtering strategies based on data type and block characteristics. This universal approach improves coding efficiency for depth data while keeping the overall system architecture manageable through systematic decision rules
Data Source
AI summary
According to the present invention, a method of processing a video signal includes the steps of: receiving the depth data corresponding to a given block containing present pixels; determining a variation of the depth data; comparing the variation of the depth data with a predetermined value; if the variation is less than the predetermined value, coding the present pixels by using a first partial filter, and if the variation is greater than the predetermined value, coding the present pixels by using a second partial filter; wherein the second partial filter is applied to a wider range than the first partial filter. Accordingly the image quality of improved; the complexity according to the filter application is reduced; and at the same time variable filtering may improve the coding efficiency.