Motion Vector Predictor Lookahead and Lookbehind for Inter Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face inefficiencies in deriving motion vector predictors, particularly in inter prediction coding, which affects compression efficiency.
Innovation Solution
The use of motion vector predictor lookahead and/or lookbehind techniques to derive motion vector predictors by summing intermediate vectors associated with reference pictures, enhancing the derivation process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional motion vector predictor derivation is used, then the device complexity is low, but the coding efficiency of inter prediction is insufficient
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing intermediate motion vectors and reference picture information before the actual motion vector prediction is needed. The decoding apparatus retrieves these pre-computed values from buffers, avoiding complex real-time calculations and reducing processing complexity while maintaining high coding efficiency through accurate motion vector prediction.
Solution Approach 2:
The patent introduces intermediary structures including intermediate motion vectors, reference picture buffers, and candidate MVP lists that mediate between the raw video data and the final motion vector prediction. These intermediaries organize and facilitate the prediction process, enabling efficient inter prediction coding without requiring direct complex analysis of the video signal.
2Measurement precision
If more reference pictures are used for motion vector prediction, then the measurement precision of motion vectors improves, but the loss of time in processing increases
Solution Approach 1:
The patent applies preliminary action by pre-computing and buffering intermediate motion vectors and reference picture data before the prediction is needed. This allows the decoding apparatus to retrieve pre-processed information quickly, achieving high measurement precision through multiple reference pictures while minimizing processing time through efficient data retrieval from pre-prepared buffers.
Solution Approach 2:
The patent ensures continuity of useful action by maintaining continuous buffers for reference pictures and intermediate motion vectors that are constantly available for prediction. This continuous availability of pre-processed data enables rapid motion vector prediction without repeated time-consuming calculations, balancing high precision with fast processing.
Data Source
AI summary
A video bitstream is received. The video bitstream includes coded information of a current block in a current picture and of a plurality of reference pictures of the current picture in a reference list. A plurality of intermediate vectors associated with one of the plurality of reference pictures is determined. The plurality of intermediate vectors includes an initial vector and a plurality of intermediate motion vectors (MVs). The initial vector is associated with the current picture. Each of the plurality of intermediate MVs is defined between two respective reference pictures of the plurality of reference pictures. A candidate motion vector predictor (MVP) is determined for the current block based on a sum of the plurality of intermediate vectors. The current block is reconstructed based on an MVP candidate list that includes the candidate MVP.