Video Coding Long-Term Reference Frame Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently utilizing long-term reference frames, leading to increased coding complexity and reduced compression speed.
Innovation Solution
A method for coding video based on a long-term reference frame is introduced, where a long-term reference frame is set according to attribute information of image frames, and a reference index is determined based on both normal and long-term reference frames for inter prediction, thereby acquiring a target matching block for efficient coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If long-term reference frame is used for inter prediction coding, then compression performance is improved, but coding complexity increases and coding speed decreases
Solution Approach 1:
The patent segments the reference frame set into short-term reference frames and long-term reference frames. The long-term reference frame is selectively used only when spatial redundancy exceeds temporal redundancy, avoiding unnecessary complexity in cases where short-term frames are sufficient. This segmentation allows the system to benefit from long-term reference frames when needed while maintaining simplicity when not needed.
Solution Approach 2:
The patent dynamically changes the parameter of reference frame selection based on redundancy evaluation. By comparing spatial redundancy and temporal redundancy parameters, the system adapts its coding strategy, using long-term reference frames only when the parameter comparison indicates superior performance, thus balancing compression gain against coding complexity.
2Manufacturing precision
If long-term reference frame is used for inter prediction coding, then compression performance is improved, but coding speed decreases
Solution Approach 1:
The patent applies partial action by using long-term reference frames only partially - specifically when spatial redundancy is greater than temporal redundancy. This selective application avoids the full computational overhead of long-term reference frame processing in all cases, thereby maintaining coding speed while achieving compression performance improvement when it matters most.
Solution Approach 2:
The system dynamically changes coding parameters based on redundancy analysis. When temporal redundancy is high, the system switches to short-term reference frames for faster coding. When spatial redundancy dominates, it switches to long-term reference frames for better compression. This parameter adaptation resolves the speed-performance tradeoff.
3Productivity
If redundancy evaluation is performed to select reference frames, then coding efficiency is improved, but computation overhead increases
Solution Approach 1:
The patent extracts only the essential redundancy information needed for decision-making, rather than performing exhaustive analysis. By focusing on comparing spatial and temporal redundancy metrics directly, the system avoids unnecessary computation while still achieving efficient reference frame selection. This extraction approach maintains coding efficiency while limiting computation overhead to what is truly necessary.
Data Source
AI summary
Disclosed is a method for coding a video based on a long-term reference frame. The method includes: setting a long-term reference frame according to attribute information of an image frame; determining a reference index of a to-be-coded image frame based on a normal reference frame and the long-term reference frame; and acquiring a target matching block by performing inter prediction based on the reference index, so as to achieve coding of the to-be-coded image frame; wherein setting the long-term reference frame according to the attribute information of the image frame includes: setting the long-term reference frame based on a degree of redundancy in a time domain and a degree of redundancy in a spatial domain of the image frame.