Rectangular Video Block Prediction for Flexible Image Decoding

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video coding technologies, such as ITU-T H.265 and JEM, lack flexibility in handling arbitrary rectangular video blocks, leading to suboptimal partitioning and coding efficiency.

Innovation Solution

Implement a video coding method that generates predictive sample values by averaging horizontal and vertical interpolations for each sample within a rectangular video block, allowing for more flexible partitioning using quadtree binary tree structures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional video coding standards (ITU-T H.265, JEM) are used for partitioning video blocks, then the coding structure is simple and standardized, but the flexibility in handling arbitrary rectangular video blocks is limited and partitioning efficiency is suboptimal

Engineering Contradiction:

Improveflexibility in handling arbitrary rectangular video blocksVSAvoidpartitioning structure complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The video block partitioning is segmented into multiple levels using quadtree and binary tree structures. The quadtree divides the video block into four quadrants recursively, while the binary tree further subdivides selected regions. This multi-level segmentation enables flexible representation of arbitrary rectangular blocks while maintaining a systematic approach to partitioning complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The partitioning structure is made dynamic by allowing selective application of quadtree and binary tree divisions based on content characteristics. The encoder can adaptively choose which regions to partition and which partitioning method to use, transforming the rigid conventional structure into a flexible, content-adaptive system that handles arbitrary rectangular blocks efficiently.

Inventive Principle:

Principle #15Dynamics

2Productivity

If more flexible partitioning structures are implemented to handle arbitrary rectangular video blocks, then coding efficiency is improved, but the device complexity and processing overhead increase

Engineering Contradiction:

Improvecoding efficiencyVSAvoidpartitioning processing complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The partitioning decisions are made in advance during the encoding process, with the quadtree and binary tree structures pre-defining the partitioning hierarchy. By establishing the partitioning structure beforehand and using systematic division rules, the encoder reduces real-time processing complexity while maintaining high coding efficiency for arbitrary rectangular blocks.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The partitioning structure uses nested quadtree and binary tree divisions where smaller partition units are nested within larger ones. This nested hierarchy allows efficient representation of arbitrary rectangular blocks at multiple scales, improving coding efficiency while managing complexity through a structured, hierarchical approach rather than treating all partitions independently.

Inventive Principle:

Principle #7Nested doll (Nesting)

Data Source

PatentUS12593066B2Image decoding apparatus and image coding apparatus

Publication Date: 2026.03.31 SHARP KK

AI summary

A method of generating a prediction for an area of video data, comprises Receiving a rectangular video block including sample values, and for each sample included in the video block, generating a predictive sample value by averaging a horizontal interpolation and a vertical interpolation corresponding to the location of the sample within the video block.