VVC Motion Vector Resolution Skipping for Faster Inter Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current acceleration algorithms for Adaptive Motion Vector Resolution (AMVR) in versatile video coding still significantly impact coding efficiency and performance, necessitating further acceleration improvements.
Innovation Solution
A method and apparatus that perform prediction at a first resolution, determine a rate-distortion cost, construct a prediction motion vector list, and skip motion estimation and compensation when certain conditions are met, using a prediction motion vector as an actual motion vector to accelerate AMVR.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion estimation and motion compensation are performed at all four resolutions during inter prediction, then the motion information expression accuracy is improved, but the time overhead increases significantly
Solution Approach 1:
The patent performs motion estimation and motion compensation at lower resolutions (1/4 pixel and 1/2 pixel) first to obtain preliminary motion vectors. These preliminary results are then used to guide the selection and refinement at higher resolutions (integer and 4 times integer pixel resolutions), avoiding the need to perform exhaustive search at all resolutions from scratch.
Solution Approach 2:
The patent divides the motion estimation process into multiple resolution stages. Each stage processes motion information at a specific resolution level, with the results from coarser resolutions informing the finer resolution processing. This segmentation allows the system to capture both large-scale and fine-grained motion characteristics efficiently.
2Productivity
If the current acceleration algorithm is applied to AMVR, then some acceleration is achieved, but coding efficiency and coding performance are still greatly affected
Solution Approach 1:
The patent dynamically adjusts the motion estimation strategy based on the current coding unit's characteristics and the results from previous resolution levels. The search range, search precision, and even the decision to skip certain resolution levels are made adaptive rather than fixed, allowing the system to maintain high performance while achieving acceleration.
Solution Approach 2:
The patent selectively applies motion estimation at different resolution levels based on whether it is necessary for the current coding unit. For some coding units, full four-resolution processing is performed; for others, fewer resolution levels are sufficient. This partial action approach maintains coding performance where needed while achieving acceleration where possible.
Data Source
AI summary
The application provide a processing method based on versatile video coding, including: performing prediction at a first resolution for a current coding unit and determining a rate-distortion cost as a current best rate-distortion cost; and performing prediction at another resolution according to following steps: constructing a prediction motion vector list of a current traversal resolution based on a reference image; determining a start search point of the current traversal resolution based on the prediction motion vector list, and determining a first rate-distortion cost; when the first rate-distortion cost and the current best rate-distortion cost meet a preset condition, skipping motion estimation and motion compensation at the current traversal resolution, and using a prediction motion vector as an actual motion vector to calculate a second rate-distortion cost of the current traversal resolution; updating the current best rate-distortion cost based on the second rate-distortion cost and the current best rate-distortion cost.