Scalable Video Coding Top-Down Prediction for Compression Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing scalable video coding methods, such as bottom-up encoding, result in suboptimal compression efficiency for enhancement layers due to uneven distribution of coding efficiency losses and increased complexity, which is not efficiently addressed by current solutions like joint BL-EL optimization.
Innovation Solution
A method that generates predictive coding parameters for a second layer representation using previously reconstructed pixels, performs prediction to generate a first layer prediction signal, and codes the predictive coding parameters and residual for the first layer representation, while optionally using inter-layer prediction and downsampling to enhance compression efficiency without increasing complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If bottom-up encoding is used for scalable video coding, then the base layer can be efficiently coded independently, but the enhancement layer suffers from suboptimal compression efficiency and uneven distribution of coding efficiency losses
Solution Approach 1:
The patent inverts the traditional bottom-up encoding approach by performing prediction for the base layer using parameters and reconstructed pixels from the enhancement layer. This top-down prediction approach allows the enhancement layer to be coded first with optimal efficiency, while the base layer benefits from prediction signals derived from the higher resolution layer, thereby improving overall compression efficiency without sacrificing independent base layer coding capability
Solution Approach 2:
The patent introduces inter-layer prediction as an additional dimension to the traditional intra-layer prediction. By utilizing spatial and temporal correlations across different resolution layers (from enhancement layer to base layer), the system creates a multi-dimensional prediction framework that enhances compression efficiency for both layers simultaneously
2Productivity
If joint base layer-enhancement layer optimization is applied, then enhancement layer compression efficiency improves, but computational complexity increases
Solution Approach 1:
The patent performs preliminary action by generating predictive coding parameters and reconstructing pixels for the enhancement layer first, before using them to predict the base layer. This preliminary processing of the enhancement layer provides ready-to-use prediction signals for the base layer, avoiding the need for complex joint optimization computations while achieving similar or better compression efficiency
Solution Approach 2:
The patent uses copying by deriving base layer prediction parameters from enhancement layer parameters through downsampling and interpolation. Instead of performing complex joint optimization, the system copies and adapts the enhancement layer's predictive coding parameters (motion vectors, coding modes) for use in base layer prediction, significantly reducing computational complexity
3Manufacturing precision
If high resolution encoding is performed for all layers, then video quality improves, but memory consumption and computational complexity increase
Solution Approach 1:
The patent applies local quality by using high-resolution reconstructed pixels from the enhancement layer specifically for generating prediction signals in the base layer, rather than maintaining high resolution throughout the entire encoding process. This localized use of high-resolution data improves video quality where needed (in prediction accuracy) while avoiding the memory overhead of storing and processing high-resolution data for all layers
Solution Approach 2:
The patent extracts only the necessary high-resolution information from the enhancement layer (predictive coding parameters and reconstructed pixels) and uses it for base layer prediction. By extracting and utilizing only the essential high-resolution components rather than processing complete high-resolution layers, the system achieves good video quality with reduced memory consumption
Data Source
AI summary
A scalable video coding method for coding a video sequence (10) results in a bit stream (20} partitionable into two layer representations (201, 2O2) of the video sequence (10). The second layer representation (2O2) has an enhanced resolution compared to the first layer representation (2O1.). The method includes generating (si) predictive coding parameters for the second layer using previously reconstructed pixels (14) in the second layer. The method further includes performing (s2) prediction for the first layer, using the predictive coding parameters and the previously reconstructed pixels (14) in the second layer, to generate a first layer prediction signal. The method further includes generating (s3) a first layer residual, based on the difference between the original image and the first layer prediction signal. The predictive coding parameters and the first layer residual are coded (s4) as the first layer representation (2O1). An encoder (30) and computer program are also provided.