Binary Mask 2D Run-Length Coding with Context-Based Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video codecs struggle with efficient compression of binary masks, particularly in regions of interest, leading to high bitstream overhead and suboptimal performance.
Innovation Solution
A method involving 2D run length encoding with context-based coding, utilizing unsigned run length and signed relative position contexts, to compress binary masks by converting them into a list of numbers and associated contexts, followed by entropy encoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If conventional video codecs are used to compress binary masks, then the coding process is simple, but the bitstream overhead is high and compression efficiency is poor
Solution Approach 1:
The binary mask is segmented into multiple rows, and each row is processed independently using run-length encoding. This segmentation allows the complex compression task to be broken down into simpler unit operations, reducing the overall complexity while achieving efficient compression through the cumulative effect of processing multiple segments.
Solution Approach 2:
The patent transforms the binary mask from a 2D pixel-based representation to a 1D run-length encoded sequence of numbers. This parameter transformation changes the fundamental representation of the data, converting spatial information into sequential run-length pairs, which significantly reduces the amount of data needed to represent the mask while maintaining the original structure information.
2Productivity
If binary masks are compressed using traditional methods, then the implementation is straightforward, but compression efficiency in regions of interest is suboptimal
Solution Approach 1:
The patent applies run-length encoding to transform the 2D binary mask into a 1D sequence of run-length pairs. This dimensional transformation compresses the data by representing contiguous regions of identical values as single entries, significantly improving compression efficiency particularly in regions with large uniform areas such as regions of interest in video frames.
Solution Approach 2:
The binary mask is pre-processed by converting it into a run-length encoded format before further compression or processing. This preliminary transformation prepares the data in an optimized representation that facilitates more efficient coding, especially for regions of interest, by preorganizing the information into meaningful run-length sequences rather than raw pixel values.
Data Source
AI summary
An apparatus including at least one processor, and at least one non-transitory memory storing instructions that, when executed with the at least one processor, cause the apparatus to perform: converting a mask into a list of numbers and at least one associated context; and encoding a plurality of the list of numbers using a context-based coding based upon the at least one associated context.