Overcomplete Basis Transform Video Coding Residual Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression methods, particularly those using Discrete Cosine Transform (DCT), face challenges at low bit rates with noticeable distortion and low compression ratios, and existing overcomplete transform-based systems are inefficient due to ad-hoc design and slow processing.
Innovation Solution
An overcomplete basis transform-based motion residual frame coding method that uses a Residual Energy Segmentation Algorithm (RESA) and Progressive Elimination Algorithm (PEA) to decompose residual images into atoms, with adaptive quantization and quadtree-based coding to efficiently reduce image size and energy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If DCT-based coding is used for motion residual pictures, then the coding process is simple and fast, but visual quality deteriorates with noticeable distortion and block artifacts at low bit rates
Solution Approach 1:
The patent changes the transform domain parameters from DCT to overcomplete basis functions (wavelets, contourlets, ridgelets, curvelets), which provide better sparsity representation for image features. This parameter change enables superior visual quality at low bit rates while maintaining computational feasibility through optimized decomposition algorithms.
Solution Approach 2:
The patent employs a composite approach by combining multiple basis functions (wavelets, contourlets, ridgelets, curvelets) into an overcomplete dictionary, allowing the system to leverage the strengths of each basis type for different image features, thereby achieving both high visual quality and efficient compression.
2Manufacturing precision
If overcomplete basis transform is used for motion residual coding, then visual quality and compression performance improve, but computational complexity increases and processing speed decreases
Solution Approach 1:
The patent segments the overcomplete transform process into distinct decomposition and coding stages, using specialized algorithms (RESA, PEA) for each stage. This segmentation allows optimization of each component independently, reducing overall computational complexity while maintaining the benefits of overcomplete basis transforms.
Solution Approach 2:
The patent performs preliminary actions by pre-defining overcomplete basis dictionaries and pre-processing residual images through motion compensation before transform coding. This preliminary preparation reduces the computational burden during the actual encoding process, improving processing speed without sacrificing visual quality.
3Ease of manufacture
If ad-hoc design is used for matched basis position coding and quantization, then implementation is straightforward, but compression performance is not optimized
Solution Approach 1:
The patent introduces feedback mechanisms in the quantization process, where quantization parameters are adaptively adjusted based on the statistical properties of transform coefficients and rate-distortion optimization. This feedback-driven approach optimizes compression performance while maintaining systematic implementation structure.
Solution Approach 2:
The patent employs dynamic quantization and coding strategies that adapt to the content being encoded. Quantization step sizes, basis function selections, and coding parameters are dynamically adjusted based on local image characteristics, achieving optimal compression performance across diverse video content.
Data Source
AI summary
The present invention provides a method to compress digital moving pictures or video signals based on an overcomplete basis transform using a modified Matching Pursuit algorithm. More particularly, this invention focuses on the efficient coding of the motion residual image, which is generated by the process of motion estimation and compensation. A residual energy segmentation algorithm (RESA) can be used to obtain an initial estimate of the shape and position of high-energy regions in the residual image. A progressive elimination algorithm (PEA) can be used to reduce the number of matching evaluations in the matching pursuits process. RESA and PEA can speed up the encoder by many times for finding the matched basis from the pre-specified overcomplete basis dictionary. Three parameters of the matched pattern form an atom, which defines the index into the dictionary and the position of the selected basis, as well as the inner product between the chosen basis pattern and the residual signal. The present invention provides a new atom position coding method using quad tree like techniques and a new atom modulus quantization scheme. A simple and efficient adaptive mechanism is provided for the quantization and position coding design to allow a system according to the present invention to operate properly in low, medium and high bit rate situations. These new algorithm components can result in a faster encoding process and improved compression performance over previous matching pursuit based video coders.