Video Coding Template Matching for Multiple Hypothesis Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding techniques face challenges in efficiently utilizing multiple hypothesis prediction (MHP) due to high signaling overhead and bandwidth requirements for determining weights in the prediction process.
Innovation Solution
Implementing template matching (TM) in the MHP process to determine weights based on template matching costs, ordering weights with higher probabilities at lower indices, reducing the need for signaling information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple hypothesis prediction (MHP) is implemented with traditional weight signaling methods, then prediction accuracy is improved, but signaling overhead and bandwidth requirements increase
Solution Approach 1:
The patent uses template matching to create a reference template from the current block's boundary samples, then compares this template against reference blocks to generate prediction hypotheses. The template acts as a copy or representation of the current block's characteristics, enabling weight determination without direct signaling of weight values.
Solution Approach 2:
The patent changes the parameter representation method by using template matching costs instead of directly signaling weight parameters. The weight is determined based on the matching cost between the template and reference blocks, transforming the problem from parameter signaling to parameter-derived determination.
2Measurement precision
If multiple hypothesis prediction (MHP) is implemented with traditional weight signaling methods, then prediction accuracy is improved, but bandwidth requirements increase
Solution Approach 1:
The template serves as a copy of the current block's boundary characteristics, which is then used to determine weights through matching operations. This approach eliminates the need to transmit weight information separately, reducing bandwidth consumption while maintaining prediction accuracy.
Solution Approach 2:
The system determines weights autonomously by performing template matching operations locally at the decoder side. The template matching process self-generates the weight information needed for MHP without requiring external signaling, making the system self-sufficient and bandwidth-efficient.
3Productivity
If weights are sorted by probability with lower indices having higher probability, then signaling efficiency is improved, but processing complexity increases
Solution Approach 1:
The patent performs preliminary sorting of prediction hypotheses based on their estimated probability or matching cost before the actual prediction process. By pre-ordering the hypotheses with the most likely candidates first, the system increases the probability that fewer bits are needed to signal the selected hypothesis index, improving signaling efficiency.
Solution Approach 2:
The patent changes the ordering parameter of hypotheses based on probability or matching cost metrics. This reparameterization of the hypothesis list allows for more efficient signaling by aligning the index values with the likelihood of selection, reducing the expected number of bits required for signaling.
Data Source
AI summary
A method of encoding or decoding video data includes determining, for a multiple hypothesis prediction (MHP) process, a plurality of prediction templates based on a plurality of weights; comparing the plurality of prediction templates to a current template of a current block; determining a weight from the plurality of weights based on the comparing of the plurality of prediction templates to the current template of the current block; determining one or more prediction hypotheses; determining a prediction signal for the current block based on the one or more prediction hypotheses and the weight; and encoding or decoding the current block based on the prediction signal.