Video Prediction With Adaptive Motion Vector List Length

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video coding technologies lack a feasible method to determine the maximum length of a candidate motion vector list in subblock merge mode, which affects encoding/decoding performance.

Innovation Solution

A method and apparatus for determining the maximum length of a candidate motion vector list in subblock merge mode by parsing specific indicators from a bitstream, considering the presence or absence of advanced temporal and planar motion vector prediction modes, and using predefined formulas to calculate the maximum length based on these indicators.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If subblock merge mode is used in inter prediction, then prediction accuracy is improved, but the complexity of determining candidate motion vector list length increases

Engineering Contradiction:
Improveprediction accuracyVSAvoidcomplexity of determining candidate motion vector list length
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces a parameter `sps_sbtmvp_enabled_flag` that can be configured to enable or disable advanced temporal motion vector prediction. By changing this parameter's value, the system can adapt the candidate motion vector list length determination between two modes: using the full list when the flag is enabled, or using a reduced list when disabled. This parameter change approach resolves the contradiction by allowing flexible adjustment of complexity based on prediction accuracy requirements.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent implements a dynamic determination method where the candidate motion vector list length is not fixed but can be adjusted based on the `sps_sbtmvp_enabled_flag` parameter. The system dynamically selects between different list lengths (full list or reduced list) depending on the prediction mode and configuration, making the complexity adaptive rather than static.

Inventive Principle:
Principle #15Dynamics

2Reliability

If advanced temporal motion vector prediction mode is enabled, then prediction performance is improved, but the bitstream parsing complexity increases

Engineering Contradiction:
Improveprediction performanceVSAvoidbitstream parsing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent performs preliminary action by parsing the `sps_sbtmvp_enabled_flag` parameter before determining the candidate motion vector list length. This flag is extracted from the bitstream in advance, allowing the system to pre-determine which determination method to use (full list or reduced list) based on the predicted picture type and the flag's value, thereby simplifying the subsequent parsing and processing steps.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent segments the bitstream parsing process into distinct steps: first parsing the `sps_sbtmvp_enabled_flag`, then using this information to determine the appropriate candidate motion vector list length. This segmentation separates the complexity of parsing the flag from the complexity of determining the list length, making the overall process more manageable and efficient.

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If candidate motion vector list length is increased, then prediction accuracy is improved, but encoding/decoding efficiency decreases

Engineering Contradiction:
Improveprediction accuracyVSAvoidencoding/decoding efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent uses the `sps_sbtmvp_enabled_flag` parameter to change the candidate motion vector list length between different values. When the flag is enabled and the predicted picture type requires it, the system uses a longer list for improved accuracy. When the flag is disabled or the picture type doesn't require it, the system uses a shorter list to maintain efficiency. This parameter-driven change resolves the contradiction by making list length adaptive rather than fixed.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies partial action by using a reduced candidate motion vector list length in certain cases (when `sps_sbtmvp_enabled_flag` is disabled or when prediction accuracy isn't critical). Instead of always using the full list, the system selectively applies the full list only when necessary, thereby avoiding excessive computational complexity while maintaining sufficient prediction accuracy for most cases.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS20260059095A1Video picture prediction method and apparatus
Publication Date: 2026.02.26 HUAWEI TECH CO LTD
  • US20260059095A1 patent drawing
  • US20260059095A1 patent drawing
  • US20260059095A1 patent drawing

AI summary

A video picture prediction method and apparatus are provided, to provide a manner of determining a maximum length of a candidate motion vector list corresponding to a subblock merge mode. The method includes: parsing a first indicator from a bitstream; if the first indicator indicates that a candidate mode used to inter predict the to-be-processed block includes an affine mode, parsing a second indicator from the bitstream, where the second indicator is used to indicate a maximum length of a first candidate motion vector list, and the first candidate motion vector list is constructed for the to-be-processed block, a subblock merge prediction mode is used for the to-be-processed block; and determining the maximum length of the first candidate motion vector list based on the second indicator.