Interpolation Filter Support for Sub-Pixel Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital video encoding and decoding techniques face challenges in accurately generating predictive data, particularly during fractional interpolation, which affects compression efficiency and video quality.
Innovation Solution
The implementation of filtering techniques that utilize a twelve-pixel filter support, coefficient symmetry, and pixel symmetry to enhance predictive data accuracy, reduce data transmission requirements, and improve interpolation at both sub-pixel and integer pixel locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional interpolation filters are used for sub-pixel resolution, then computational complexity is reduced, but predictive data accuracy deteriorates
Solution Approach 1:
The patent applies asymmetry by allowing non-symmetric filter coefficients in the interpolation filter. Instead of using symmetric filters that simplify computation, the patent uses asymmetric filters with independently determined coefficients that better match the actual image content characteristics, thereby improving predictive data accuracy for sub-pixel resolution while managing complexity through selective application.
Solution Approach 2:
The patent changes the parameters of the interpolation filter by adapting filter coefficients based on local image characteristics such as variance and gradient. This allows the filter to adjust its behavior dynamically across different regions of the image, improving predictive accuracy for both smooth and textured areas while maintaining reasonable computational complexity through parameter adaptation rather than full complexity filters.
2Measurement precision
If adaptive filter coefficients are used for each sub-pixel location, then interpolation accuracy is improved, but data transmission requirements increase
Solution Approach 1:
The patent applies local quality by determining filter coefficients adaptively for different local regions of the image based on local characteristics such as variance and gradient. Instead of using a single global filter, the patent adjusts filter parameters locally to match the specific content characteristics of each region, improving interpolation accuracy while reducing the amount of data needed compared to full adaptive filtering.
Solution Approach 2:
The patent implements partial adaptation by applying adaptive filtering selectively based on local image characteristics. Rather than adapting all filter coefficients everywhere, the patent applies adaptation only where needed (e.g., in regions with high variance or specific gradient patterns), reducing the overall data transmission requirements while maintaining interpolation accuracy in critical regions.
3Productivity
If fractional motion vectors are used for sub-pixel locations, then compression efficiency is improved, but computational complexity increases
Solution Approach 1:
The patent applies preliminary action by performing adaptive filtering and coefficient determination during the encoding process before compression. By pre-computing and storing the necessary filter parameters and adaptive coefficients during encoding, the patent enables efficient decoding with fractional motion vectors without requiring complex computations during the compression stage, thus improving overall compression efficiency while managing computational complexity.
Data Source
AI summary
This disclosure describes filtering techniques applied by an encoder and a decoder during the prediction stage of a video encoding and/or decoding process. The filtering techniques may enhance the accuracy of predictive data used during fractional interpolation, and may improve predictive data of integer blocks of pixels. There are several aspects to this disclosure, including a useful twelve-pixel filter support that may be used for interpolation, techniques that use coefficient symmetry and pixel symmetry to reduce the amount of data needed to be sent between an encoder and a decoder to configure the filter support for interpolation, and techniques for filtering data at integer pixel locations in a manner that is similar to sub-pixel interpolation. Other aspects of this disclosure concern techniques for encoding information in the bitstream to convey the type of filter used, and possibly the filter coefficients used. Predictive coding of filter coefficients is also described.