Decoder-Side Motion Vector Derivation for Selective Block Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards face challenges in efficiently handling higher resolution videos due to increased bandwidth demands and computational complexity, particularly in decoder-side motion vector derivation processes.
Innovation Solution
Implementing a method for selective enablement of decoder-side motion vector derivation (DMVD) tools using merge modes and motion vector differences, with refinements based on coding information, and processor-executable code for video processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If decoder-side motion vector derivation (DMVD) tool is enabled for all blocks, then motion information refinement accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent changes the parameter of DMVD tool applicability from universal to conditional by introducing syntax elements that indicate whether DMVD is enabled for specific blocks. This allows the system to adapt the level of motion information refinement based on block characteristics, thereby improving accuracy where needed while reducing computational complexity where not required.
Solution Approach 2:
The patent introduces dynamic control of DMVD tool usage through syntax elements that can be set differently for different blocks. This dynamic approach allows the decoder to selectively apply motion vector derivation refinement based on coding information, achieving a balance between refinement accuracy and computational load across different video blocks.
2Device complexity
If DMVD tool is selectively enabled based on coding information, then computational complexity is reduced, but motion information refinement accuracy may deteriorate
Solution Approach 1:
The patent applies local quality by making DMVD tool applicability block-specific rather than uniform across the entire video stream. By using syntax elements to indicate DMVD enablement on a per-block basis, the system provides high motion information refinement accuracy only where needed, while reducing computational complexity in blocks where refinement is less critical.
Solution Approach 2:
The patent implements partial action by selectively applying DMVD tool to only those blocks that require motion information refinement. This partial application approach avoids the excessive computational complexity of applying DMVD to all blocks while ensuring that blocks benefiting from refinement receive the necessary processing.
3Loss of energy
If merge mode with motion vector differences is used, then bandwidth usage is reduced, but decoder-side processing complexity increases
Solution Approach 1:
The patent introduces syntax elements as intermediaries that carry information about DMVD tool applicability from the encoder to the decoder. These syntax elements enable the decoder to understand which blocks require DMVD processing without transmitting full motion vector information, thus reducing bandwidth usage while providing necessary guidance for selective processing.
Data Source
AI summary
Devices, systems and methods for digital video coding, which includes restrictions on decoder-side motion vector derivation based on coding information, are described. An exemplary method for video processing includes making a decision, for a conversion between a current block of a video and a bitstream representation of the video, regarding a selective enablement of a decoder side motion vector derivation (DMVD) tool for the current block, the DMVD tool deriving a refinement of motion information signaled in the bitstream representation, and the conversion using a merge mode and motion vector differences that are indicated by a motion direction and a motion magnitude, and performing, based on the decision, a conversion between the current block and the bitstream representation.