Intra-Inter Combined Prediction for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies limit prediction performance by treating intra prediction and inter prediction independently, failing to leverage the advantages of both methods effectively.
Innovation Solution
A combined prediction method for P frames or B frames that adaptively selects and weights intra prediction blocks and inter prediction blocks based on rate-distortion optimization, using weighting coefficients inversely proportional to prediction distortion statistics to improve prediction precision and coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If intra prediction and inter prediction are used independently, then the coding structure is simple and easy to implement, but the prediction performance is limited and cannot fully utilize the advantages of both methods
Solution Approach 1:
The patent merges intra prediction block and inter prediction block through weighted averaging to form a combined prediction block. The encoder performs both intra and inter prediction, then combines them using weighting coefficients to achieve better prediction accuracy than either method alone, while maintaining a relatively simple coding structure.
Solution Approach 2:
The patent introduces weighting coefficients as adjustable parameters to control the contribution of intra and inter prediction blocks. By dynamically changing these weights based on prediction distortion statistics, the system optimizes prediction accuracy without significantly increasing structural complexity.
2Productivity
If only one prediction method (intra or inter) is selected by encoder decision, then the coding process is simple, but the prediction performance is limited and cannot comprehensively consider the advantages and disadvantages of both methods
Solution Approach 1:
The patent makes the prediction method dynamic by using rate-distortion optimization to adaptively select whether to apply combined prediction for each coding unit. The system dynamically adjusts based on local characteristics, transmitting a 1-bit flag to indicate whether combined prediction is used, thereby optimizing coding efficiency while improving prediction precision where beneficial.
3Measurement precision
If combined prediction method is used for all coding units, then prediction precision is improved, but the coding complexity increases and bitrate increases due to transmitting additional flags and inter prediction information
Solution Approach 1:
The patent applies combined prediction selectively rather than universally. A 1-bit flag is transmitted for each coding unit to indicate whether combined prediction is used, allowing the system to apply the more complex method only where it provides benefit, thereby balancing prediction precision improvement with coding complexity and bitrate constraints.
4Measurement precision
If combined prediction method is used for all coding units, then prediction precision is improved, but the bitrate increases due to transmitting additional inter prediction information
Solution Approach 1:
The patent transmits inter prediction information and 1-bit flags only for coding units where combined prediction is determined to be beneficial through rate-distortion optimization. This selective transmission reduces the additional bitrate overhead compared to universal application, while still achieving prediction precision improvement where needed.
Data Source
AI summary
An intra-frame and inter-frame combined prediction method for P frames or B frames. The method comprises: self-adaptively selecting by means of a rate-distortion optimization (RDO) decision whether to use the intra-frame and inter-frame combined prediction or not; using a method for weighting an intra prediction block and an inter prediction block in the intra-frame and inter-frame combined prediction to obtain a final prediction block; and obtaining the weighting coefficient of the intra prediction block and the inter prediction block according to prediction distortion statistics of the prediction method. Therefore, prediction precision can be improved, and coding and decoding efficiency of the prediction blocks are improved. The advantages of intra prediction and inter prediction are fully utilized in the present invention; and the optimal prediction parts of the two methods are selected to be combined, so that to a certain extent, areas with excessive distortion can be removed out of the intra prediction block and the inter prediction block, thus obtaining a better prediction effect and achieving excellent practicality and robustness.