HMVP Candidate List Construction for Video Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding technologies face challenges in efficiently encoding and decoding digital video due to high bandwidth demands, particularly as the number of connected devices increases, necessitating improved compression techniques to manage growing bandwidth requirements.
Innovation Solution
The implementation of a method for constructing candidate lists for motion vectors using pre-defined relative positions, including spatial and temporal candidates, and history-based motion vector prediction, which enhances the encoding and decoding processes by incorporating look-up tables to store and update motion information for improved prediction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If traditional video compression techniques are used, then bandwidth consumption is high, but compression efficiency is insufficient
Solution Approach 1:
The patent applies preliminary action by constructing a candidate list of motion vectors before the actual encoding process. The candidate list includes spatial candidates from neighboring blocks and temporal candidates from reference frames, prepared in advance to guide the motion estimation process and improve compression efficiency while reducing bandwidth consumption
Solution Approach 2:
The patent uses copying by reusing motion vector information from previously encoded blocks (spatial candidates from neighboring blocks and temporal candidates from reference frames) instead of encoding completely new motion vectors, thereby improving compression efficiency and reducing bandwidth requirements
2Measurement precision
If more motion vector candidates are checked, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies local quality by differentiating between spatial candidates (from locally neighboring blocks) and temporal candidates (from reference frames), assigning different priorities and checking conditions to each type. This allows the system to achieve good prediction accuracy by focusing computational resources on the most relevant candidates while managing complexity
Solution Approach 2:
The patent uses parameter changes by dynamically adjusting the number of candidates to check based on the relationship between the total number of candidates and a threshold value K. When the difference exceeds K, fewer candidates are checked; when it's within K, more candidates are checked, allowing the system to adapt prediction accuracy to computational constraints
Data Source
AI summary
A method of video decoding includes checking one or more video blocks in pre-defined relative positions compared to a first video block to derive a first set of candidates, wherein each of the first set of candidates is derived from corresponding motion information of the video blocks checked, checking, upon a determination that a number of the first set of candidates has not reached a predetermined limit, one or multiple tables to derive a second set of candidates, and processing the first video block based on a bitstream representation of the first video block and the constructed candidate list.