Multi-Hypothesis Video Decoding with Derived Bi-Prediction Weights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional multi-hypothesis prediction (MHP) signaling in video coding techniques suffer from high signaling overhead and processing requirements, and there is a need for improved prediction accuracy for motion vector predictors and prediction blocks.
Innovation Solution
The proposed techniques enhance MHP by reducing signaling overhead and processing requirements through improved weight determination and additional inter-prediction modes, utilizing weighted superposition of multiple motion-compensated prediction signals, and iterative accumulation of prediction blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional multi-hypothesis prediction (MHP) signaling is used in video coding, then prediction accuracy for motion vector predictors and prediction blocks is improved, but signaling overhead and processing requirements increase
Solution Approach 1:
The patent extracts and separates the weight determination process from the conventional MHP signaling. By determining weights based on motion vector differences and reference picture indices rather than signaling them explicitly, the patent removes the overhead of weight signaling while maintaining prediction accuracy. This extraction of the weight determination mechanism resolves the contradiction between improved prediction accuracy and reduced signaling overhead.
Solution Approach 2:
The patent enables the decoding device to self-determine prediction weights using locally available information (motion vector differences, reference picture indices) without requiring additional signaling from the encoder. This self-service approach allows the system to maintain high prediction accuracy while eliminating the need for explicit weight signaling, thus resolving the technical contradiction.
2Measurement precision
If conventional multi-hypothesis prediction (MHP) signaling is used in video coding, then prediction accuracy for motion vector predictors and prediction blocks is improved, but processing requirements increase
Solution Approach 1:
The patent extracts the weight determination logic from complex MHP signaling and replaces it with simpler calculations based on motion vector differences and reference picture indices. This extraction reduces the processing burden while maintaining prediction accuracy, resolving the contradiction between improved precision and reduced processing requirements.
Solution Approach 2:
The patent changes the parameters used for weight determination from explicitly signaled values to derived values based on motion vector differences and reference picture indices. This parameter change simplifies the processing required for MHP while maintaining or improving prediction accuracy, thus resolving the technical contradiction.
3Measurement precision
If bi-prediction mode with different weights is used for current block, then prediction accuracy is improved, but signaling overhead increases
Solution Approach 1:
The patent extracts the weight information from explicit signaling and derives it from motion vector differences and reference picture indices. This extraction eliminates the need to signal weight values separately, reducing signaling overhead while maintaining the ability to apply different weights for improved prediction accuracy in bi-prediction mode.
Data Source
AI summary
An example device for decoding video data includes one or more processors configured to: determine that a first weight and a second weight are specified for a bi-prediction mode predicted current block of video data; determine whether the current block is to be predicted using multi-hypothesis prediction (MHP) mode with the bi-prediction mode as a base mode; in response to determining that the current block is to be predicted using the MHP mode with the bi-prediction mode as the base mode, determine an additional inter-prediction mode of the MHP mode; generate a first prediction block according to the bi-prediction mode; generate a second prediction block according to the additional inter-prediction mode; generate a final prediction block for the current block according to the MHP mode using the first prediction block and the second prediction block; and decode the current block using the final prediction block.