Video Coding Consistency via Content-Based Partition Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video coding and decoding processes are 'lossy' and often result in perceptible errors, particularly when different coding techniques are applied to similar content portions within a video sequence, leading to dissatisfactory viewing experiences due to varying artifacts.
Innovation Solution
A method that partitions a video sequence based on content similarity, groups partitions with similar features, and assigns consistent coding parameters to each group to minimize artifacts, ensuring consistent presentation of similar content portions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If different coding techniques are applied to various portions of video sequence, then coding efficiency is improved, but artifact consistency deteriorates
Solution Approach 1:
The patent applies different coding techniques to different portions of the video sequence based on local content characteristics. By analyzing content similarity between portions and selectively applying coding techniques, the system achieves both coding efficiency improvement and artifact consistency within similar content regions.
2Adaptability or versatility
If similar content portions are coded differently, then coding flexibility is improved, but viewing experience deteriorates
Solution Approach 1:
The system analyzes content similarity and applies coding techniques adaptively based on local content characteristics. This ensures that similar content portions receive consistent coding treatment, maintaining reliable viewing experience while preserving necessary coding flexibility for different content types.
3Adaptability or versatility
If coding parameters vary across video portions, then compression adaptability is improved, but perceptible errors increase
Solution Approach 1:
The patent implements content-based analysis to identify similar portions and applies consistent coding parameters within those portions. This local quality approach maintains compression adaptability across different video content while minimizing perceptible errors by ensuring consistent coding within similar content regions.
Data Source
AI summary
The present disclosure describes techniques for coding video data in a manner that provides consistency to portions of the video that have similar content. According to such techniques, a video sequence may be parsed into partitions and content of the partitions may be analyzed. Partitions may be grouped together based on detected similarities in content. Coding parameters may be selected for each partition based on the partition's membership in the groups. Thus, when the video sequence is coded, coding parameters for frames of two commonly-grouped partitions may be similar, which causes coded video data to have similar presentation.