Layer-Based Video Encoding With Masked Block Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video codecs face challenges in balancing legibility and compression ratio, particularly when encoding video signals with detailed content like small text, as aggressive quantization can degrade legibility while less aggressive quantization compromises compression efficiency.
Innovation Solution
The technique generates multiple layers and corresponding masks for each block of a video signal, applying lossy compression to each layer and transmitting the compressed layers and masks to the decoder, allowing for more aggressive compression and improved image quality by distinguishing relevant pixels from irrelevant ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the encoder quantizes the DCT coefficients aggressively to reduce compression artifacts and improve legibility, then image quality and legibility are improved, but compression ratio deteriorates
Solution Approach 1:
The patent divides each block into multiple sub-blocks, allowing independent quantization of each sub-block. This segmentation enables the encoder to apply different quantization strengths to different regions, improving overall image quality while maintaining compression efficiency by avoiding uniform aggressive quantization across the entire block.
Solution Approach 2:
The patent applies different quantization parameters to different sub-blocks within a block, allowing local optimization of image quality in regions requiring higher precision (such as text areas) while maintaining stronger compression in regions where it can be applied without sacrificing quality.
2Manufacturing precision
If the encoder uses lossless compression for synthetic content like text, then legibility is improved, but compression efficiency deteriorates
Solution Approach 1:
The patent identifies synthetic content regions (such as text) within blocks and applies appropriate quantization parameters specifically to those regions, allowing lossless or near-lossless compression for text while applying more aggressive compression to natural image regions, thus maintaining both legibility and compression efficiency.
Solution Approach 2:
By segmenting blocks into sub-blocks, the patent can identify and isolate synthetic content regions, allowing different compression strategies to be applied to different segments - lossless compression for text-containing sub-blocks and lossy compression for background sub-blocks.
Data Source
AI summary
A technique for encoding a video signal generates multiple layers and multiple corresponding masks for each of a set of blocks of the video signal. Each of the layers for a given block is a rendition of that block, and each of the masks distinguishes pixels of the respective layer that are relevant in reconstructing the block from pixels that are not. The encoder applies lossy compression to each of the layers and transmits the lossily compressed layers and a set of the masks to a decoder, such that the decoder may reconstruct the respective block from the layers and the mask(s).