RDPCM Quantization for Video Residual Compression

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding techniques face inefficiencies in compressing video data, particularly in handling residual data through traditional quantization methods that are not optimized for spatial signals, leading to suboptimal bandwidth usage and coding efficiency.

Innovation Solution

The implementation of Residue Differential Pulse Code Modulation (RDPCM) with uniform quantization and adaptive quantization step sizes and offsets for different pixel locations, applied to intra- and inter-prediction residual blocks, to reduce the amount of data needed to be signaled, enhancing bandwidth efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If traditional quantization methods are used for residual data, then the coding process is simple, but the bandwidth efficiency and coding efficiency are suboptimal

Engineering Contradiction:
Improvebandwidth efficiencyVSAvoidcoding process complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The patent applies uniform quantization with adaptive quantization step sizes and offsets tailored to different pixel locations and spatial characteristics of residual data. This parameter optimization improves bandwidth efficiency by better matching the quantization process to the actual distribution and importance of residual values, achieving more efficient compression without fundamentally changing the quantization framework

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent implements location-dependent quantization where different quantization step sizes and offsets are applied to different pixel locations based on their spatial characteristics. This local optimization allows more important or variable regions to be preserved with higher fidelity while applying stronger compression to less critical areas, thereby improving overall bandwidth efficiency

Inventive Principle:
Principle #3Local quality

2Loss of information

If uniform quantization RDPCM is applied to residual blocks, then the amount of data to be signaled is reduced, but the quantization may introduce more distortion

Engineering Contradiction:
Improvedata amountVSAvoidresidual data precision
Core Design Contradiction:
Loss of informationVSManufacturing precision

Solution Approach 1:

The patent employs adaptive quantization step sizes and offsets that are adjusted based on the spatial characteristics and statistical properties of the residual data. By dynamically tuning these parameters, the system achieves better preservation of critical information while still reducing the overall data amount, balancing compression efficiency with precision retention

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

Different quantization parameters are applied to different regions of the residual block based on local characteristics such as variance, gradient, or importance metrics. This ensures that regions requiring higher precision maintain better accuracy despite quantization, while other regions can be more aggressively compressed, thus reducing total data while minimizing overall distortion

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP3120544B1Quantization processes for residue differential pulse code modulation
Publication Date: 2018.06.06 QUALCOMM INC
  • EP3120544B1 patent drawingFigure 1
  • EP3120544B1 patent drawingFigure 2
  • EP3120544B1 patent drawingFigure 3

AI summary

In an example, system and method for coding encoded video data. A quantized residue differences block associated with a current block of a picture is generated and a reconstructed prediction residue for each quantized residue difference in the block of quantized residue differences is generated. Each reconstructed prediction residue is added to its corresponding original prediction value to produce a current block of a picture. Generating the reconstructed prediction residue includes performing inverse quantization on blocks of quantized residue differences, wherein the inverse quantization reconstructs data to which uniform quantization residue differential pulse code modulation (RDPCM) has been applied and wherein the quantization is of the form: ΔX q = floor ((X+ αQ)/Q) where Q is quantization step and α is quantization offset.