Post-Filter Rescaling for Sub-Pel Video Sample Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding systems face challenges in accurately correcting reconstructed samples due to sub-pixel precision issues, leading to inefficiencies in compression and transmission of digital video signals.

Innovation Solution

A device determines a scaling parameter and offset based on sub-pel correction precision and classification, applying a scaled correction to reconstructed samples to enhance precision, using a computer program product for implementation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If sub-pel correction precision is used for reconstructed samples, then correction accuracy is improved, but computational complexity and processing overhead increase

Engineering Contradiction:
Improvecorrection precisionVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies different correction precisions to different regions or types of samples based on classification. By categorizing samples and applying appropriate correction levels to each class, the system achieves high precision where needed while reducing processing complexity for other samples, thus resolving the contradiction between correction accuracy and computational burden.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent dynamically adjusts correction parameters based on sample characteristics and classification results. By changing correction parameters adaptively rather than applying uniform high-precision correction to all samples, the system maintains high correction accuracy for critical samples while reducing overall processing complexity.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If scaling parameter precision is increased, then correction accuracy is improved, but data transmission and storage requirements increase

Engineering Contradiction:
Improvescaling parameter precisionVSAvoiddata volume
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent uses classification to identify which samples require high-precision scaling parameters and which can use lower precision. By applying high precision only where necessary, the system improves correction accuracy for critical samples while minimizing the overall data volume required for storing and transmitting scaling parameters.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent applies high-precision scaling parameters to only a subset of samples that benefit most from increased precision, rather than applying high precision universally. This partial application of high precision achieves sufficient correction accuracy while significantly reducing data transmission and storage requirements.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4661393A1Classification-based post-filter rescaling
Publication Date: 2025.12.10 INTERDIGITAL CE PATENT HOLDINGS SAS
  • EP4661393A1 patent drawingFigure 1A
  • EP4661393A1 patent drawingFigure 1B
  • EP4661393A1 patent drawingFigure 1C

AI summary

Systems, methods, and instrumentalities are disclosed herein for a device that may be configured to determine a correction of a reconstructed sample at sub-pel precision. The correction may be associated with a sub-pel correction precision. The device may determine a scaling parameter associated with the reconstructed sample. The scaling parameter may be associated with a scaling parameter precision. The device may determine a scaled correction based on the correction, the scaling parameter, and/or the scaling parameter precision. The device may apply the scaled correction to the reconstructed sample.