Sign Data Hiding in Video Coding Using Single Variable Threshold

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current sign data hiding techniques in video coding require complex criteria for determining when to apply the process, such as storing two variables for each coefficient group, and only allow hiding one sign per group, limiting their effectiveness.

Innovation Solution

Proposed techniques simplify the decision-making process by using a single variable, such as the position or number of non-zero coefficients, to determine whether to apply sign data hiding, and allow hiding multiple signs within the parity of the sum of absolute values, enabling more efficient coding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If complex criteria with multiple variables are used to determine sign data hiding, then the accuracy of sign hiding determination may improve, but the device complexity and processing overhead increase

Engineering Contradiction:
Improvesign hiding determination accuracyVSAvoidvariable storage and processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and uses only the most critical variable (position of the last non-zero coefficient in scanning order) to determine sign data hiding, discarding the need for multiple variables and complex criteria while maintaining effective sign hiding performance

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the coefficient group processing by using a simple threshold-based decision (comparing the position of the last non-zero coefficient to a threshold value) to divide coefficient groups into those that undergo sign data hiding and those that do not, simplifying the overall process

Inventive Principle:
Principle #1Segmentation

2Ease of manufacture

If only one sign is hidden per coefficient group, then the implementation simplicity is maintained, but the coding efficiency and rate-distortion performance are limited

Engineering Contradiction:
Improveimplementation simplicityVSAvoidcoding efficiency
Core Design Contradiction:
Ease of manufactureVSProductivity

Solution Approach 1:

The patent applies sign data hiding to multiple signs within a coefficient group when the condition (position of last non-zero coefficient exceeding threshold) is met, rather than limiting to a single sign, thereby achieving better compression efficiency while maintaining a simple implementation framework

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent makes the sign data hiding process universal by allowing multiple signs to be hidden within a single coefficient group under the same simple condition, enabling the mechanism to handle various coding scenarios more effectively

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS9313498B2Sign hiding techniques for quantized transform coefficients in video coding
Publication Date: 2016.04.12 QUALCOMM INC
  • US9313498B2 patent drawing
  • US9313498B2 patent drawing
  • US9313498B2 patent drawing

AI summary

This disclosure proposes techniques for encoding and decoding transform coefficients in a video coding process. In particular, this disclosure proposes techniques determining whether or not to apply a sign data hiding process for a group of transform coefficients, and techniques for applying the sign data hiding process. In one example, this disclosure describes a method for decoding video data comprising determining a block of transform coefficients, determining whether to perform a sign data hiding process for at least one transform coefficient in the block of transform coefficients based on a single variable compared to a threshold, and decoding sign information for the block based on the determination of whether to perform the sign data hiding process.