HEVC Intra Prediction Computational Load Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

High Efficiency Video Coding (HEVC) techniques face significant computational complexity during the 3D-HEVC intra prediction process, limiting the practical application of 3D-HEVC encoders and decoders due to the need for extensive rate distortion (RD) cost calculations.

Innovation Solution

A method that generates a full rate distortion calculation list of selected intra coding modes, disabling segment-wise depth coding mode for each intra prediction mode, and comparing the RD cost with and without this mode to select the most efficient prediction unit, thereby reducing unnecessary calculations and computational load.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If all possible intra coding modes are evaluated using full rate distortion calculation, then coding efficiency is improved, but computational complexity increases extremely

Engineering Contradiction:
Improvecoding efficiencyVSAvoidcomputational complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the intra coding modes into two categories: depth intra prediction modes (which model depth relationships) and conventional intra prediction modes (which handle texture). This segmentation allows the encoder to apply different evaluation strategies to each category, reducing the overall computational burden while maintaining coding efficiency.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial rate distortion calculation instead of full rate distortion calculation for depth intra prediction modes. This partial action approach evaluates only the necessary components (depth model parameters) rather than performing complete RD optimization, significantly reducing computational complexity while preserving essential coding performance.

Inventive Principle:
Principle #16Partial or excessive action

2Manufacturing precision

If segment-wise depth coding mode is enabled for all intra prediction modes, then coding accuracy is improved, but unnecessary calculations increase computational load

Engineering Contradiction:
Improvecoding accuracyVSAvoidencoding speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent applies segment-wise depth coding selectively only to depth intra prediction modes where it is applicable and beneficial, rather than forcing it on all intra prediction modes. This local quality approach ensures that the complex segmentation and modeling operations are performed only where they provide value, avoiding unnecessary calculations for modes where depth modeling is not appropriate.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent performs preliminary evaluation to determine whether segment-wise depth coding should be applied before executing the full coding process. By pre-assessing the suitability and potential benefit of SDC for each depth intra prediction mode, the encoder可以避免 performing unnecessary complex calculations when SDC would not improve results.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3140988B1Method and device for reducing a computational load in high efficiency video coding
Publication Date: 2019.09.11 HUAWEI TECH CO LTD
  • EP3140988B1 patent drawingFigure 1
  • EP3140988B1 patent drawingFigure 2
  • EP3140988B1 patent drawingFigure 3

AI summary

A method (500) for reducing a computational load in high efficiency video coding includes generating (504) a full rate distortion calculation list of selected intra coding modes where the intra coding modes including intra prediction modes and depth modeling modes. A rate distortion cost is determined, with a segment-wise depth coding mode being disabled, for each intra prediction mode in the full rate distortion calculation list and a smallest rate distortion cost intra prediction mode is selected (506). A rate distortion cost for a particular intra prediction mode is calculated with the segment-wise depth coding mode enabled (510). After comparison (512), one of the particular intra prediction mode (514) and the smallest rate distortion cost intra prediction mode (516) having the smallest rate distortion cost is applied to a prediction unit.