Two-Stage Intra Search for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding systems face challenges in maintaining high performance and quality while adhering to silicon area and power constraints, especially with the increasing number of codecs and the need for real-time HD and 4K encoding on limited processing devices.
Innovation Solution
The implementation of a two-stage intra search method with a split search architecture, including an open-loop first stage and a closed-loop second stage, along with fine-grained rate distortion estimation enhancements, allows for dynamic sizing of coding units and mixed prediction and transform types, optimizing compression efficiency and quality across multiple codecs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single-stage intra search is used, then the device complexity is reduced, but the compression efficiency and quality deteriorate
Solution Approach 1:
The intra search process is divided into two distinct stages: a first stage that performs a limited search with fewer candidates, and a second stage that performs a more thorough search on selected candidates. This segmentation allows the encoder to balance computational complexity with compression efficiency by not examining all possible candidates in a single exhaustive search, thus resolving the contradiction between device complexity and compression efficiency.
2Loss of substance
If high compression efficiency is achieved, then the bit rate is reduced, but the encoding time increases
Solution Approach 1:
The first stage of the intra search performs preliminary evaluation of multiple candidates to identify the most promising ones before the second stage conducts a more detailed search. This preliminary action filters out poor candidates early, allowing the encoder to achieve high compression efficiency through thorough search of selected candidates while avoiding the time cost of exhaustive search across all possible candidates, thus resolving the contradiction between bit rate reduction and encoding time.
3Manufacturing precision
If exhaustive intra search is performed, then the quality is improved, but the power consumption increases
Solution Approach 1:
The exhaustive search is segmented into two stages where the first stage quickly evaluates many candidates to identify a limited set of promising ones, and the second stage performs detailed quality optimization only on these selected candidates. This segmentation maintains high video quality through thorough search of relevant candidates while significantly reducing power consumption compared to a single exhaustive search, resolving the contradiction between video quality and power consumption.
Data Source
AI summary
Example apparatus to encode video disclosed herein include an encoder to perform an intra search first stage based on source pixels of a source video frame to determine first intra candidates to predict a block of the source video frame. In disclosed examples, the encoder is also to perform an intra search second stage based on reconstructed pixels of neighboring blocks associated with the first intra candidates to determine a second intra candidate. In disclosed examples, the encoder is further to encode the block of the source video frame based on the second intra candidate.