Video Compression via Non-Linear Quantization and Modular Arithmetic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video data communication between multimedia Systems on Chip (SoC) and off-chip memory requires high bandwidth, increasing implementation costs due to large bandwidth demands, necessitating a method for efficient video compression that maintains video quality while reducing component costs.
Innovation Solution
The method employs non-linear quantization and modular arithmetic computation for differential coding of video data blocks, generating a codeword that achieves compression efficiency and reduces data communication and component costs by using a video differential coder and codeword generator.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video data is stored and communicated between off-chip memory and multimedia SoC, then video data access is enabled, but bandwidth requirement increases and implementation cost increases
Solution Approach 1:
The patent extracts only the essential video data information by performing differential coding that captures only the differences between reference and current blocks. This extraction approach reduces the quantity of data that needs to be stored and communicated, thereby reducing bandwidth requirements while maintaining video data access capability
Solution Approach 2:
The patent applies non-linear quantization to transform the differential values into a compact representation. By changing the parameter representation from raw pixel differences to quantized differential codes, the data volume is significantly reduced, enabling efficient storage and communication with lower bandwidth requirements
2Reliability
If video data is stored and communicated between off-chip memory and multimedia SoC, then video data access is enabled, but implementation cost of off-chip memory and SoC memory interface increases
Solution Approach 1:
The patent extracts only the essential video data information by performing differential coding that captures only the differences between reference and current blocks. This extraction approach reduces the quantity of data that needs to be stored and communicated, thereby reducing bandwidth requirements and the cost of memory interfaces
Solution Approach 2:
The patent applies non-linear quantization to transform the differential values into a compact representation. By changing the parameter representation from raw pixel differences to quantized differential codes, the data volume is significantly reduced, enabling efficient storage and communication with lower implementation costs
3Quantity of substance
If compression is applied to video data, then compression ratio improves and bandwidth requirement reduces, but video quality may deteriorate
Solution Approach 1:
The patent applies different processing approaches to different parts of the video data: differential coding is applied locally to capture block differences, and non-linear quantization is applied to prioritize the representation of important differential values. This local quality approach maintains video quality in critical areas while achieving compression
Solution Approach 2:
The patent applies non-linear quantization to transform the differential values into a compact representation. By changing the parameter representation from raw pixel differences to quantized differential codes, the data volume is significantly reduced while maintaining perceptual video quality through the non-linear transformation that preserves important visual information
Data Source
AI summary
A system and method for video compression utilizes non-linear quantization and modular arithmetic computation to perform differential coding on multiple blocks of video data and uses a result of the differential coding to generate a codeword.