Hybrid Video Prediction Blocks Combining Inter and Intra Modes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression techniques face challenges in achieving high performance due to the large amount of data in video streams, requiring efficient methods to reduce bandwidth and storage requirements while maintaining video quality.
Innovation Solution
The method involves combining inter-prediction and intra-prediction within a block using a template, where inter-predicted pixels from a reference frame and intra-predicted pixel values are used to form a prediction block, which is then used to encode or decode the current block, improving rate/distortion measures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If inter-prediction using motion vectors is used for video compression, then bandwidth requirements are reduced, but prediction accuracy deteriorates for certain block types
Solution Approach 1:
The patent combines inter-prediction and intra-prediction methods into a hybrid prediction mode. The prediction block is divided into two portions: the first portion uses inter-predicted pixels from reference frames, while the second portion uses intra-predicted pixel values from the current frame. This merging allows the system to leverage the motion compensation strength of inter-prediction while incorporating the spatial correlation accuracy of intra-prediction, thereby maintaining lower bandwidth requirements while improving overall prediction accuracy.
Solution Approach 2:
The patent segments the prediction block into two distinct portions with different prediction methods. The first portion (corresponding to template dimensions) is generated using inter-prediction from reference frames, while the second portion is generated using intra-prediction from the current frame. This segmentation allows each portion to be optimized for its specific characteristics, improving overall prediction accuracy without significantly increasing bandwidth requirements.
2Measurement precision
If intra-prediction using pixels from the current frame is used, then prediction accuracy is improved, but bandwidth and storage requirements increase
Solution Approach 1:
The patent applies local quality by using intra-prediction only for the second portion of the prediction block where spatial correlation is more beneficial, rather than applying it to the entire block. The first portion continues to use inter-prediction from reference frames. This selective application of intra-prediction improves prediction accuracy locally where needed while avoiding the increased bandwidth and storage requirements that would result from applying intra-prediction to the entire block.
3Measurement precision
If a hybrid prediction mode combining inter and intra prediction is used, then rate/distortion performance is improved, but computational complexity increases
Solution Approach 1:
The patent implements partial action by applying intra-prediction only to a portion (the second portion) of the prediction block rather than the entire block. The first portion is handled by inter-prediction. This partial application of the more computationally complex intra-prediction method achieves improved rate/distortion performance while limiting the increase in computational complexity compared to applying intra-prediction to the entire block.
Data Source
AI summary
A prediction block is determined for a current block of a current frame of a video stream using a template having pixel locations that conform to a subset of the pixel locations of the current block. A first portion of the prediction block having the same pattern of pixel locations as the template is populated by inter-predicted pixel values, and the remaining portion of the prediction block is populated by intra-predicted pixel values. The intra-predicted pixel values may be determined using inter-predicted pixel values of the first portion, pixel values of pixels adjacent to the current block, or both.