Inter Prediction Motion Vector Rounding for Overflow-Safe Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently encoding and decoding high-resolution video data while maintaining image quality, particularly in handling motion vectors that exceed the valid bit-depth range, leading to potential arithmetic overflows and ambiguities in motion vector values.
Innovation Solution
Implement unified motion vector rounding and clipping methods for various inter prediction processes, including TMVP, ATMVP, MMVD, and affine modes, using equations to ensure motion vectors remain within the valid bit-depth range, thereby avoiding overflow and ambiguity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If motion vector rounding is performed using conventional methods for different inter prediction processes, then each process can be handled separately, but the complexity of implementing multiple different rounding methods increases hardware complexity
Solution Approach 1:
The patent applies a universal motion vector rounding method that works for multiple inter prediction processes (TMVP, ATMVP, MMVD, affine modes) using a single unified equation. This eliminates the need for separate rounding implementations for each prediction type, thereby reducing hardware complexity while maintaining support for all prediction processes.
Solution Approach 2:
The patent merges multiple different motion vector rounding operations into a single unified rounding operation. By combining the rounding logic for TMVP, ATMVP, MMVD, and affine modes into one common equation, the patent reduces the number of separate hardware circuits needed and simplifies the overall implementation.
2Quantity of substance
If motion vectors are stored without rounding control, then the full range of motion vector values can be represented, but arithmetic overflow occurs when motion vectors exceed the valid bit-depth range
Solution Approach 1:
The patent applies preliminary rounding to motion vectors before they are used in subsequent calculations. By rounding motion vectors to the valid bit-depth range early in the processing pipeline, the patent prevents arithmetic overflow from occurring later during motion compensation and other operations.
Solution Approach 2:
The patent changes the parameter representation of motion vectors by applying rounding to constrain them within the valid bit-depth range. This parameter transformation ensures that motion vectors remain within representable limits while preserving the essential motion information needed for accurate prediction.
3Manufacturing precision
If different rounding methods are used for different prediction modes, then each mode can be optimized for its specific characteristics, but ambiguities and inconsistencies arise in motion vector values
Solution Approach 1:
The patent applies local quality by tailoring the rounding equation to specifically address the needs of different inter prediction processes while maintaining a unified approach. The rounding method is locally adapted to handle the specific characteristics of TMVP, ATMVP, MMVD, and affine modes, ensuring both accuracy and consistency.
Data Source
AI summary
A method for video encoding is provided. In the method the encoder obtains a video picture that is split into multiple video blocks. The encoder determines an inter prediction process for one of the multiple video blocks. When determining that the inter prediction process is one of derivation processes for temporal motion vector prediction (TMVP), alternative temporal motion vector prediction (ATMVP), or merge mode with motion vector differences (MMVD), the encoder performs a MV rounding process to at least one MV in the inter prediction process according to a first MV rounding mode that is based on the following equation: L=(A+the_first_offset)»Shift.