Method and device for encoding video to limit bidirectional prediction and block merging, and method and device for decoding video
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video codecs face high computational complexity during bidirectional prediction, leading to increased hardware load and potential degradation of image quality due to the need for repeated motion estimation.
Innovation Solution
Limiting bidirectional motion estimation and block merging based on whether the L0 and L1 reference pictures are the same and the difference between their motion vectors, reducing unnecessary computations and maintaining image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If bidirectional motion estimation is performed for all blocks, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent changes the parameter of motion estimation application by introducing a flag (bmvl_flag) to selectively enable or disable bidirectional motion vector prediction based on specific conditions. This allows the system to adapt between full bidirectional estimation and limited estimation, resolving the contradiction between accuracy and complexity.
Solution Approach 2:
The patent applies bidirectional motion estimation selectively to specific blocks rather than uniformly across all blocks. By using conditions such as reference picture list configuration and motion vector differences, the patent applies complex bidirectional estimation only where necessary, maintaining accuracy for critical blocks while reducing overall computational load.
2Manufacturing precision
If bidirectional prediction is used, then image quality is improved, but hardware load increases
Solution Approach 1:
The patent implements partial bidirectional prediction by applying it only to blocks that meet specific criteria rather than all blocks. This partial application maintains image quality where needed while significantly reducing the hardware load associated with performing bidirectional prediction on every block in the video sequence.
3Productivity
If block merging is performed, then compression efficiency is improved, but image quality degrades
Solution Approach 1:
The patent modifies the block merging process by introducing conditional checks on reference picture lists and motion vector differences before allowing merging. This changes the merging parameter from aggressive (merge whenever possible) to selective (merge only when conditions are met), maintaining compression efficiency while preserving image quality by avoiding inappropriate merges.
Data Source
AI summary
Provided are a method and apparatus for encoding a video and a method and apparatus for decoding a video. The method of encoding includes: determining, for a current block, bidirectional motion estimation information comprising an L0 motion vector, an L1 motion vector, an L0 reference vector, and an L1 reference picture by performing bidirectional motion estimation on the current block; determining whether to use the determined bidirectional motion estimation information as prediction information of the current block based on whether the L0 reference picture and the L1 reference picture are the same and based on a difference value between the L0 motion vector and the L1 motion vector; and in response to determining to use the determined bidirectional motion estimation information as the prediction information of the current block, encoding the determined bidirectional motion estimation information as the prediction information of the current block.