Dynamic Reference Index Assignment for Video Coding Efficiency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The H.264 video compression standard experiences a deterioration in coding efficiency in scenes with large motions due to conventional reference index assignment methods, which prioritize pictures with the same parity over temporally closer reference pictures, leading to increased coded reference index amounts.

Innovation Solution

An image coding apparatus that dynamically assigns reference indices to reference pictures, giving a smaller value to the temporally closer reference picture, regardless of parity, to enhance coding efficiency and image quality in scenes with large motions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If reference indices are assigned prioritizing pictures with the same parity over temporally closer reference pictures, then coding efficiency is maintained for scenes with small motions, but coding efficiency deteriorates in scenes with large motions and the coded amount of reference indices increases

Engineering Contradiction:
Improvecoding efficiencyVSAvoidadaptability to different motion scenarios
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The reference index assignment method is made dynamic by switching between two assignment strategies based on motion magnitude detection. When large motion is detected, the system assigns smaller reference index values to temporally closer reference pictures regardless of parity. When small motion is detected, it assigns smaller values to pictures with the same parity. This dynamic adaptation resolves the contradiction by allowing the system to optimize for different motion scenarios rather than being fixed to a single assignment rule

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the assignment parameters (reference index values) based on the detected motion characteristics of the scene. By detecting motion magnitude and adjusting which parameter (temporal distance vs. parity) receives priority in the assignment algorithm, the system adapts to different coding scenarios. This parameter change approach allows optimal reference index assignment for both small and large motion scenes

Inventive Principle:
Principle #35Parameter changes

2Loss of information

If conventional reference index assignment prioritizes same-parity pictures, then coding is efficient for small motions, but the coded amount of reference index increases for large motions

Engineering Contradiction:
Improvecoded amount of reference indexVSAvoidcoding efficiency in large motion scenes
Core Design Contradiction:
Loss of informationVSProductivity

Solution Approach 1:

The system dynamically adjusts the reference index assignment strategy based on detected motion magnitude. For large motion scenes, it switches to prioritizing temporally closer reference pictures, which reduces the coded amount of reference index while maintaining coding efficiency. This dynamic switching resolves the contradiction between minimizing coded amount and maintaining efficiency in large motion scenarios

Inventive Principle:
Principle #15Dynamics

3Productivity

If reference index assignment is fixed based on parity, then implementation is simple, but coding efficiency deteriorates in large motion scenes

Engineering Contradiction:
Improvecoding efficiencyVSAvoidcomplexity of reference index assignment
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system performs preliminary motion detection and analysis before the reference index assignment process. By detecting the motion magnitude in advance and determining the appropriate assignment strategy beforehand, the system avoids complex real-time decision-making during coding. This preliminary action approach maintains implementation simplicity while achieving adaptive optimization for different motion scenarios

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The motion detection mechanism serves as an intermediary that bridges the simple fixed assignment method and the complex adaptive assignment method. It provides the necessary information about scene characteristics to select the appropriate assignment strategy, adding minimal complexity while enabling adaptive optimization without requiring the entire assignment system to be complex

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS8254451B2Image coding apparatus, image coding method, image coding integrated circuit, and camera
Publication Date: 2012.08.28 PANASONIC HOLDINGS CORP
  • US8254451B2 patent drawing
  • US8254451B2 patent drawing
  • US8254451B2 patent drawing

AI summary

An image coding apparatus includes a reference index assignment unit that assigns reference indices to reference pictures which can be referred to by a current picture to be coded, each of the reference indices having a different value and being assigned to a corresponding one of the plural reference pictures. An image coding unit performs field-coding on the current picture to be coded, by referring to the reference pictures indicated by the reference indices assigned by the reference index assignment unit, and codes the reference indices. The reference index assignment unit assigns, to one of two reference pictures included in the plural reference pictures, a reference index having a value smaller than a value of a reference index to be assigned to the other reference picture. One of the two reference pictures is temporally closer to the current picture to be coded.