Dynamic GOP Adjustment for Scene Change Adaptation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

MPEG2 video coding is sub-optimal due to fixed GOP parameters, which do not adapt to scene changes or content complexity, leading to inefficient compression and storage requirements.

Innovation Solution

A method to dynamically adjust GOP parameters by analyzing video data to align scene changes with GOP starts, extending or shortening GOP lengths, and converting B pictures to P pictures based on complexity metrics, thereby optimizing coding efficiency and reducing I frame usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If fixed GOP parameters are used in MPEG2 video coding, then the encoding process is simple and stable, but the compression efficiency is sub-optimal and does not adapt to scene changes or content complexity

Engineering Contradiction:
ImproveAdaptability to scene changesVSAvoidGOP parameter adjustment complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic GOP parameters by allowing the GOP length to vary based on detected scene changes. When a scene change is detected, the GOP length is extended to include more pictures, enabling the encoding structure to adapt dynamically to content variations rather than using a fixed GOP length throughout the video sequence.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes GOP parameters (specifically GOP length and structure) based on scene change detection and content complexity analysis. By modifying these parameters dynamically according to the video content characteristics, the system achieves better compression efficiency without requiring complete re-encoding, thus balancing adaptability with implementation feasibility.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If more I pictures are used to ensure reference quality, then the reference picture quality is high, but the encoded data size increases and processing speed decreases

Engineering Contradiction:
ImproveReference picture qualityVSAvoidProcessing speed
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies local quality enhancement by selectively extending GOP length and using more reference pictures only in regions where scene changes occur or where content complexity requires it. In stable regions, the original GOP structure is maintained, thus achieving high reference picture quality where needed without uniformly increasing I picture count throughout the entire video sequence.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent uses partial action by extending GOP length and adding reference pictures only partially - specifically at scene change boundaries and in high-complexity regions rather than throughout the entire video. This selective approach provides sufficient reference quality for problematic areas while maintaining processing efficiency in stable regions.

Inventive Principle:
Principle #16Partial or excessive action

3Quantity of substance

If B pictures are used extensively for compression, then the file size is reduced, but the coding performance deteriorates in complex scenes with scene changes

Engineering Contradiction:
ImproveEncoded data sizeVSAvoidCoding performance
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent dynamically adjusts the picture type assignment (I, P, or B) based on scene change detection and content complexity metrics. In complex scenes or at scene boundaries, the system converts B pictures to P or I pictures to maintain coding performance, while in simple scenes, B pictures are retained for maximum compression. This parameter change approach balances file size reduction with coding performance reliability.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS8837602B2Content adaptive video encoder and coding method
Publication Date: 2014.09.16 MK SYST USA INC
  • US8837602B2 patent drawing
  • US8837602B2 patent drawing
  • US8837602B2 patent drawing

AI summary

A method of pre-processing a sequence of video picture data comprising I pictures, P pictures and B pictures, where the pictures are arranged as a plurality of Group Of Pictures (GOP) and each of the Group Of Pictures has a first picture and a plurality of Sub Group of Pictures (Sub-GOP). The method uses an analysis of the sequence of video picture data to determine a position of a scene change relative to a position of a start of a new GOP within the sequence of video picture data and then moves the start of the new GOP to coincide with the scene change by extending or decreasing a current GOP length according to whether the scene change and the start of a new GOP are within a predetermined number of pictures of each other. An apparatus for carrying out the method is also provided.