Video Interpolation Filtering with Phase-Aligned 6- to 12-Tap Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies, such as VVC, face challenges with suboptimal phase characteristics and phase misalignments in interpolation filters, particularly for 6-tap, 8-tap, and 12-tap filters, which affect video quality and efficiency.
Innovation Solution
Implementing interpolation filters with specific sets of coefficient values and lengths (6, 8, 10, or 12) tailored to the characteristics of reconstructed samples, ensuring accurate phase alignment and improved frequency responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the filter length is increased to 6-tap, 8-tap, or 12-tap, then video quality and frequency response are improved, but phase misalignment and phase characteristics deteriorate
Solution Approach 1:
The patent changes the phase shift parameter of the interpolation filter based on the filter length. Specifically, it applies different phase shift values (e.g., -90 degrees for 6-tap, -180 degrees for 8-tap, -270 degrees for 12-tap) to compensate for the phase misalignment that occurs with longer filter lengths. This parameter adjustment resolves the contradiction by maintaining phase alignment accuracy while allowing the use of longer filters for improved frequency response.
2Reliability
If the filter length is increased to 12-tap, then significant video quality gains are achieved, but device complexity and computational load increase
Solution Approach 1:
The patent introduces a phase shift parameter that can be adjusted based on filter length, allowing the system to use longer filters (12-tap) for improved video quality while compensating for their inherent phase misalignment. This enables the deployment of more complex filters without sacrificing phase accuracy, effectively managing the trade-off between video quality and implementation complexity.
Data Source
AI summary
There is provided a method for encoding or decoding a video, the video comprising a sequence of pictures which includes a first picture and a second picture. The method comprises obtaining a first group of values of samples included in a first block, wherein the first block is included in the first picture. The method further comprises selecting, from a set of filters, a filter to use for generating a second group of values included in a second block, wherein the second block is included in the first picture or the second picture. The method comprises, using the selected filter and the first group of values, generating the second group of values, wherein the filter comprises a set of 6, 8, 10, or 12 coefficient values, and the set of coefficient values is selected from given groups of coefficient values.