Sub-Block Motion Candidate Selection for Lower-Complexity Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face challenges in efficiently managing motion vector prediction, particularly in sub-blocks, leading to increased bandwidth demand and computational complexity in digital video transmission.
Innovation Solution
Implementing sub-block based motion vector prediction methods, including sub-block based temporal motion vector prediction (SbTMVP) and affine coding modes, to optimize the construction of merge candidate lists and improve motion vector prediction efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If sub-block based motion vector prediction methods are implemented, then video coding efficiency is improved, but device complexity increases
Solution Approach 1:
The current block is divided into multiple sub-blocks, and motion vector prediction is performed independently for each sub-block. This segmentation allows the system to achieve better coding efficiency by capturing local motion variations while maintaining a systematic approach to complexity management through uniform processing of each sub-block.
Solution Approach 2:
Different motion vector prediction modes (TMVP, SbTMVP, affine) are selectively applied to different sub-blocks based on local motion characteristics. This allows the system to allocate computational resources dynamically, applying complex prediction methods only where needed while using simpler methods in other areas, thus improving overall coding efficiency without uniformly increasing complexity across the entire block.
2Productivity
If multiple prediction modes (TMVP, SbTMVP, affine) are supported, then motion vector prediction efficiency is improved, but processing complexity increases
Solution Approach 1:
The system dynamically selects among multiple prediction modes (TMVP, SbTMVP, affine) based on the specific characteristics of each current block and sub-block. This dynamic adaptation allows the system to achieve high prediction efficiency when complex modes are beneficial while falling back to simpler modes when appropriate, thus improving overall efficiency without requiring all modes to be processed equally in all cases.
Solution Approach 2:
The system changes the prediction mode parameter based on block characteristics, such as block size, motion complexity, and reference picture availability. By adjusting the mode selection parameter dynamically, the system optimizes prediction efficiency for different scenarios while managing processing complexity through conditional logic rather than uniform complex processing.
3Productivity
If sub-block based prediction is used, then compression ratio is improved, but bandwidth demand increases
Solution Approach 1:
The system copies motion vector information from reference blocks in previous pictures to predict current block motion. By reusing and adapting existing motion vector data through temporal motion vector prediction, the system achieves better compression without requiring additional bandwidth for transmitting new motion information, as the prediction leverages already-transmitted reference data.
Solution Approach 2:
Motion vector prediction is performed in advance using reference pictures before the actual current block encoding. This preliminary prediction allows the system to prepare motion vector estimates that reduce the amount of data needing transmission, thereby improving compression ratio while managing bandwidth demand through pre-computed prediction values.
Data Source
AI summary
Devices, systems and methods for digital video coding, which include sub-block based inter prediction methods, are described. An exemplary method for video processing includes determining, for a conversion between a current block of video and a bitstream representation of the video, a maximum number of candidates in a sub-block based merge candidate list and/or whether to add sub-block based temporal motion vector prediction (SbTMVP) candidates to the sub-block based merge candidate list based on whether temporal motion vector prediction (TMVP) is enabled for use during the conversion or whether a current picture referencing (CPR) coding mode is used for the conversion, and performing, based on the determining, the conversion.