Video Signal Encoding Using Single Optimized Graph
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video signal processing technologies face challenges in efficiently handling next-generation video content with high spatial resolution, high frame rate, and high dimensionality, requiring significant increases in memory storage, memory access rate, and processing power, particularly in combining multiple weighted graphs for effective signal processing.
Innovation Solution
A method for combining multiple weighted graphs into a single optimization weighted graph using a statistical formulation based on a maximum likelihood criterion, deriving optimal conditions for common graph transform (CGT) and common graph frequency (CGF) estimation, and searching for the closest graph Laplacian matrix to create an optimal combined graph.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple weighted graphs are combined using traditional averaging methods, then the process is simple and computationally efficient, but the modeling accuracy and coding gain are insufficient
Solution Approach 1:
The patent transforms the graph combination problem from a simple averaging operation into an optimization problem by changing the parameters being optimized. Instead of uniformly averaging graph Laplacian matrices, the method optimizes a combined graph Laplacian matrix to maximize coding gain and minimize average quadratic cost, thereby improving modeling accuracy while managing complexity through structured optimization.
Solution Approach 2:
The patent creates a composite graph structure by combining multiple weighted graphs into a single optimized graph. This composite graph integrates information from multiple source graphs (different characteristics like friendship, political viewpoint, local location) into a unified structure that captures complex relationships more effectively than individual graphs or simple averages.
2Manufacturing precision
If next-generation video content with high spatial resolution and high frame rate is processed, then the video quality and dimensionality are improved, but the memory storage, memory access rate, and processing power requirements increase tremendously
Solution Approach 1:
The patent extracts and processes only the essential relationship information from video signals by representing inter-pixel relationships in graph form. This extraction approach focuses computation on capturing the most important structural relationships rather than processing all pixel data uniformly, thereby reducing the processing power and memory requirements needed to achieve high video quality.
Solution Approach 2:
The patent applies local quality by using weighted graphs to represent local inter-pixel relationships with varying weights. Different regions and relationships are modeled with appropriate weights, allowing the system to focus computational resources on capturing locally important relationships while maintaining overall video quality, thus reducing total processing requirements.
Data Source
AI summary
The present invention provides a method for encoding a video signal by using a single optimized graph, comprising the steps of: obtaining a residual block; generating graphs from the residual block; generating an optimized graph and an optimized transform by combining the graphs, wherein the graphs are combined on the basis of an optimization step; and performing a transform for the residual block on the basis of the optimized graph and the optimized transform.