Intra Prediction Pixel Extrapolation for Better Video Compression

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video compression standards replicate unavailable pixels without considering the underlying patterns in the frame, leading to sub-optimal prediction and inefficient compression.

Innovation Solution

Generate unavailable pixels through extrapolation based on actual computed differences between available pixels, using predefined thresholds to ascertain the strength of patterns, thereby replicating patterns in the extrapolated pixels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If unavailable pixels are replicated using available pixels during intra prediction, then the prediction process can continue, but frame patterns are broken and compression efficiency deteriorates

Engineering Contradiction:

Improveprediction process continuityVSAvoidframe pattern preservation

Core Design Contradiction:

Ease of operationVSManufacturing precision

Solution Approach 1:

The patent applies preliminary action by pre-computing differences between available pixels and using these pre-calculated differences to extrapolate unavailable pixels. This allows the prediction process to maintain pattern continuity without requiring actual unavailable pixel data, resolving the contradiction between operational continuity and pattern preservation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by replicating the difference patterns observed in available pixels to generate extrapolated values for unavailable pixels. Instead of directly copying available pixels (which breaks patterns), the method copies the relational differences between pixels to maintain the underlying frame patterns while filling gaps.

Inventive Principle:

Principle #26Copying

2Productivity

If pixel replication is used to substitute unavailable pixels, then the prediction can proceed, but residual errors increase and compression efficiency decreases

Engineering Contradiction:

Improveprediction processing speedVSAvoidcompression efficiency

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The patent applies parameter changes by transforming the prediction approach from direct pixel replication to difference-based extrapolation. This changes the fundamental parameter of how unavailable pixels are generated, resulting in lower residual errors and improved compression efficiency while maintaining processing speed.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If simple pixel replication is used, then device complexity remains low, but prediction accuracy deteriorates due to ignored pixel relationships

Engineering Contradiction:

Improveprediction algorithm complexityVSAvoidpixel prediction accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces the mechanical system of direct pixel copying with a computational system that calculates and applies difference relationships. This substitution increases prediction accuracy by considering pixel relationships while keeping the implementation relatively simple through mathematical operations rather than complex structural changes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12621433B2Prediction-based extrapolation of pixels for improved video compression

Publication Date: 2026.05.05 XILINX INC

AI summary

Methods and systems for generating missing reference pixels for intra prediction of coding units are described. A pattern amongst a plurality of available reference pixel samples from a set of reference pixel samples is computed. The pattern can be determined based on a computed difference between actual pixel values of the available reference pixel samples. The patterns are learned based on a comparison of the computed difference between the actual pixel values to a predetermined threshold. The unavailable pixel values are then generated based on the learned pattern. Further, one or more image effects corresponding to the available reference pixel samples are automatically replicated in the generated pixels as well.