Video Decoding Resolution Adjustment for Directional Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video codecs face challenges in maintaining high video playback quality under limited mobile bandwidth, as they often use the same resolution for encoding and decoding, leading to distortion and inability to perform directional prediction when resolutions differ between video blocks.
Innovation Solution
A method where the electronic device adjusts the reconstruction block of a reference block to match the resolution of a current block, allowing for directional prediction even when resolutions differ, by either upsampling or downsampling the reference block pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the same resolution is used for encoding and decoding video frames, then the encoding process is simple, but the video playback quality deteriorates under limited bandwidth due to distortion and inability to perform directional prediction
Solution Approach 1:
The patent applies parameter changes by adjusting the resolution of reference blocks to match the current block's resolution. When the current block has a higher resolution than the reference block, the reference block is upscaled; when it has a lower resolution, it is downscaled. This resolution parameter adjustment enables directional prediction to be performed accurately across different resolutions, improving video playback quality while maintaining encoding feasibility
Solution Approach 2:
The patent introduces dynamic resolution adjustment where the reference block resolution is not fixed but adaptively changed based on the current block's resolution requirements. This dynamic approach allows the system to switch between different resolution levels and scaling operations depending on the specific encoding context, enabling directional prediction to work effectively under varying resolution conditions
2Reliability
If different resolutions are used for encoding and decoding different video blocks, then video playback quality improves, but directional prediction cannot be performed due to resolution mismatch
Solution Approach 1:
The patent introduces an intermediary resolution adjustment process that mediates between reference blocks and current blocks with different resolutions. By inserting a resolution matching step (upsampling or downsampling) between the reference block and the current block, directional prediction can be performed on matched-resolution blocks, and the result is then applied to the original current block, thus enabling directional prediction across resolution differences
Solution Approach 2:
The patent applies preliminary action by adjusting the resolution of reference blocks before performing directional prediction. The resolution matching operation is performed in advance on the reference block so that when directional prediction is executed, both the reference block and current block are at the same resolution, enabling the prediction algorithm to work correctly without resolution mismatch issues
Data Source
AI summary
A video decoding method is disclosed and performed by an electronic device, the method including: obtaining a first resolution used in decoding a current block and a second resolution used in decoding a reference block in a current video frame, the reference block being a reference block used for performing directional prediction on the current block in the current video frame; building a reconstruction block from the reference block; when the first resolution is greater than the second resolution, adjusting the reconstruction block of the reference block to the first resolution to obtain a first reconstruction block; and performing directional prediction on the current block according to the first reconstruction block and the current block.