Video Coding Reference Field Selection Adaptation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video compression algorithms, such as those in the H.264 standard, often fail to utilize the best reference field for coding, leading to suboptimal video quality due to restrictive conventions that limit reference field selection to like-for-like fields, even when a more suitable field is temporally closer or of better quality.

Innovation Solution

A method and apparatus that dynamically select between the first and second fields of a reference frame based on the content of the video signal, specifically by marking frames as 'still type' or 'moving type' based on motion between top and bottom fields, allowing the best suited reference field to be chosen for coding, thereby maximizing redundancy removal.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If like-for-like field selection is used (top field references top field, bottom field references bottom field), then decoder complexity is reduced and implementation is simplified, but video quality deteriorates when motion exists between fields

Engineering Contradiction:
Improvedecoder implementation simplicityVSAvoidvideo coding quality
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The patent applies dynamics by making the reference field selection adaptive rather than static. The encoder dynamically determines whether to use like-for-like field selection or cross-field selection based on detected motion between fields. This allows the system to switch between different reference field selection modes depending on the content characteristics, optimizing video quality while maintaining decoder simplicity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameter of reference field selection based on motion detection. When motion between top and bottom fields is detected, the system changes from like-for-like selection to cross-field selection. This parameter change allows the encoder to adapt to different video content types (static vs. moving scenes) and maintain high video quality without increasing decoder complexity.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If cross-field reference selection is used to improve video quality in motion scenarios, then video quality improves, but encoder complexity increases due to additional motion analysis

Engineering Contradiction:
Improvevideo coding qualityVSAvoidencoder complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies partial action by performing motion detection only where necessary. Instead of analyzing all possible reference fields exhaustively, the system performs a simple motion check between top and bottom fields of the current frame and selects reference fields based on this partial analysis. This reduces encoder complexity while still capturing the essential motion information needed for optimal reference field selection.

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent uses self-service by having the encoder automatically detect motion characteristics and select appropriate reference fields without requiring complex external control or manual intervention. The motion detection and reference field selection are integrated into the encoding process itself, allowing the system to adapt to content characteristics autonomously without adding significant complexity.

Inventive Principle:
Principle #25Self-service

3Manufacturing precision

If multiple reference fields are always evaluated to ensure optimal selection, then video quality is maximized, but computational complexity and processing time increase

Engineering Contradiction:
Improvereference field selection accuracyVSAvoidencoding speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent applies segmentation by dividing the reference field selection process into distinct stages: first evaluating like-for-like fields, then conditionally evaluating cross-field options only when motion is detected. This segmented approach avoids the need to always evaluate all possible reference fields, reducing computational complexity while maintaining optimal selection accuracy when needed.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses preliminary action by performing a quick motion detection check before proceeding to full reference field evaluation. This preliminary assessment allows the encoder to quickly determine whether cross-field reference selection is necessary, avoiding unnecessary computational overhead in static scenes while preparing for optimal selection in motion scenarios.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2865181A1Apparatus and method for coding a video signal
Publication Date: 2015.04.29 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

AI summary

An apparatus and method are provided for coding a video signal, wherein each picture frame of the video signal comprises one or more corresponding reference frames, each reference frame R comprising a first field (for example a top field or a bottom field) and a second field (for example a bottom field or top field). A current frame to be coded is received, and a first field or a second field of a reference frame selected for coding a first field of the current frame. The selecting is performed based on the video content of the reference frame. The first field of the current frame is coded using the selected field of the reference frame.