Video Encoding Signaling Cost Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video encoders face computational inefficiencies due to the need to test numerous coding parameters, leading to wasted cycles and potential video quality losses from inaccurate cost estimates, especially in estimating the rate and distortion components of encoding costs.
Innovation Solution
A method to reduce computational cycles by evaluating the signaling cost of tool settings relative to a reference cost, where settings with signaling costs greater than a percentage of the reference cost are skipped, and those with lower costs are fully evaluated, optimizing the selection of the lowest encoding cost option.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If all available tools and their corresponding settings are tested to maximize coding efficiency, then video quality is improved, but computational complexity increases significantly
Solution Approach 1:
The patent applies partial action by evaluating only a subset of tool settings rather than all possible settings. It calculates signaling cost estimates and selectively evaluates settings based on whether their signaling cost is below a threshold (e.g., 50% of reference cost), performing full evaluation only for promising candidates while skipping others to reduce computational complexity
Solution Approach 2:
The patent performs preliminary action by calculating signaling cost estimates for all tool settings before conducting full evaluation. This preliminary assessment identifies which settings are worth evaluating in detail, allowing the system to filter out high-cost settings early and focus computational resources on low-cost candidates that are more likely to be selected
2Productivity
If tool settings are eliminated based on inaccurate cost estimates, then computational cycles are reduced, but video quality losses occur
Solution Approach 1:
The patent uses feedback by comparing the signaling cost estimate against a reference cost threshold to make decisions about whether to evaluate a setting. This feedback mechanism allows the system to adaptively select which settings to evaluate based on their estimated cost, ensuring that settings likely to be selected are fully evaluated while avoiding unnecessary evaluation of high-cost settings
Solution Approach 2:
The patent changes the parameter being evaluated from comprehensive quality assessment to signaling cost estimation as a filtering criterion. By using signaling cost as a preliminary parameter to gatekeeper full evaluation, the system can eliminate settings that are unlikely to be selected without needing to perform expensive quality assessments on all settings
3Reliability
If computational resources are spent on tool settings that are not utilized, then encoding completeness is maintained, but computational waste increases
Solution Approach 1:
The patent extracts the signaling cost calculation from the full evaluation process and uses it as a separate filtering stage. By taking out this preliminary assessment step, the system can identify and exclude settings that are unlikely to be selected before investing computational resources in full evaluation, thereby eliminating wasted computational cycles on settings that would not be utilized
Data Source
AI summary
A system that includes at least one memory and circuitry coupled to the at least one memory, wherein the circuitry is to access media from the at least one memory, wherein the circuitry is to: select one or more settings to apply to encode the video based on a cost of signaling applicable settings.