Macroblock-Level Adaptive Quantization for Video Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video compression techniques fail to optimally manage bitrate and perceptual quality at the macroblock level, as they do not account for varying visual information distribution within frames, leading to inefficient compression and potential distortion.

Innovation Solution

A macroblock-level quality-aware video optimization method that adjusts quantization parameters (QP) based on empirical testing and visual characteristics, allowing for fine control of quality degradation and bitrate reduction, while maintaining controllable perceptual quality, by employing a video optimizer system that decodes, adjusts QP values, and re-encodes video streams on a macroblock basis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If frame-level QP modification is used to maintain average bitrate, then bitrate control is improved, but visual quality in detail-rich regions deteriorates

Engineering Contradiction:
ImprovebitrateVSAvoidvisual quality
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent divides the video frame into multiple macroblocks and applies different QP values to different macroblocks based on their visual complexity. This segmentation allows fine-grained control of quantization strength in different regions, preserving visual quality in detail-rich areas while achieving bitrate reduction in flat regions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements local quality adjustment by computing visual complexity metrics for each macroblock and assigning QP values accordingly. Regions with high visual complexity (edges, textures, motion) receive lower QP values to preserve quality, while flat regions receive higher QP values to reduce bitrate, creating locally optimized quality distribution.

Inventive Principle:
Principle #3Local quality

2Device complexity

If uniform QP is applied across all macroblocks, then device complexity is reduced, but compression efficiency deteriorates

Engineering Contradiction:
Improveencoding complexityVSAvoidcompression efficiency
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent introduces dynamic QP adjustment where the QP value for each macroblock is determined by real-time computation of visual complexity metrics. This dynamic approach adapts the quantization strength to the local content characteristics, significantly improving compression efficiency compared to static uniform QP while maintaining manageable computational complexity through efficient metric computation.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The encoding system automatically adjusts QP values based on self-computed visual complexity metrics without requiring external input or manual configuration. The system serves itself by using the visual content characteristics to drive the quantization decisions, eliminating the need for complex external control mechanisms.

Inventive Principle:
Principle #25Self-service

3Quantity of substance

If higher QP values are used to reduce bitrate, then loss of information increases, but perceptual quality degradation becomes unacceptable

Engineering Contradiction:
ImprovebitrateVSAvoidperceptual quality
Core Design Contradiction:
Quantity of substanceVSLoss of information

Solution Approach 1:

The patent applies different quantization strengths to different regions based on their visual importance. Detail-rich regions maintain lower QP values to preserve perceptual quality, while flat regions use higher QP values for aggressive compression. This local differentiation ensures that information loss occurs primarily in imperceptible areas while preserving quality in visually critical regions.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the QP parameter dynamically based on visual complexity metrics. By computing metrics such as variance, gradient magnitude, or edge density for each macroblock, the system adapts the quantization parameter to match the local content characteristics, optimizing the balance between bitrate reduction and perceptual quality preservation.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2719178B1Macroblock-level adaptive quantization in quality-aware video optimization
Publication Date: 2022.02.16 CITRIX SYSTEMS INC
  • EP2719178B1 patent drawingFigure 1
  • EP2719178B1 patent drawingFigure 2
  • EP2719178B1 patent drawingFigure 3

AI summary

A method of optimizing an encoded video stream comprising one or more video frames, each video frame comprising a plurality of macroblocks, each macroblock comprising a plurality of pixels. The method includes receiving an encoded macroblock, decoding the encoded macroblock, and extracting a first quantization parameter. The first quantization parameter corresponds to quantization settings originally used for compressing the encoded macroblock. The method also includes computing a second quantization parameter based at least in part on the first quantization parameter, re-encoding the decoded macroblock based on the second quantization parameter, and providing the re-encoded macroblock.