Superpixel Border Encoding with Shared-Contour Chain Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image and video coding techniques face challenges in efficiently encoding superpixels with discontinuities, leading to increased reconstruction artifacts and higher coding bitrates due to the need to encode shared borders between regions multiple times.
Innovation Solution
A new chain coding strategy that encodes superpixels borders without selecting a starting reference vertex per region, using symbols that produce a skewed probability distribution to optimize entropy coding, allowing each region border to be encoded only once and reducing the overall coding bitrate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional chain codes encode each superpixel border independently, then complete border information is captured, but the coding bitrate increases due to redundant encoding of shared borders
Solution Approach 1:
The patent merges the encoding of shared borders by having adjacent superpixels reference the same border symbols. When two superpixels share a border, the border is encoded once and both superpixels reference it, eliminating redundant encoding while preserving complete border information.
Solution Approach 2:
Border symbols serve multiple functions by being referenced by multiple adjacent superpixels. A single border encoding can be reused by multiple superpixels that share that border, making the encoding process universal rather than duplicate.
2Ease of manufacture
If chain codes use uniform symbol distribution, then encoding is straightforward, but compression efficiency decreases
Solution Approach 1:
The patent changes the symbol distribution parameter from uniform to non-uniform by introducing reference symbols that point to previously encoded borders. This creates a skewed probability distribution where certain symbols (references) appear more frequently, enabling better entropy coding compression while maintaining encoding simplicity through the reference mechanism.
3Ease of operation
If a starting reference vertex is selected for each superpixel, then border encoding is systematic, but device complexity increases due to vertex selection and management
Solution Approach 1:
The patent extracts the vertex selection complexity by removing the requirement to select starting reference vertices for each superpixel. Instead, borders are encoded in a systematic order (e.g., raster scan order) without needing to identify or manage starting vertices, simplifying the encoding process while maintaining systematics.
4Measurement precision
If superpixel borders are encoded multiple times for each adjacent region, then each region gets accurate border representation, but the number of symbols increases
Solution Approach 1:
Instead of copying the entire border encoding for each adjacent superpixel, the patent uses reference symbols that point to the original border encoding. Each superpixel gets accurate border representation by referencing the original encoding, without duplicating the actual border symbols.
Data Source
AI summary
The present invention relates to a method for encoding the borders of pixel regions of an image, wherein the borders contain a sequence of vertices subdividing the image into regions of pixels (superpixels), by generating a sequence of symbols from an alphabet including the step of: defining for each superpixel a first vertex for coding the borders of the superpixel according to a criterion common to all superpixels; defining for each superpixel the same coding order of the border vertices, either clockwise or counter-clockwise; defining the order for coding the superpixels on the base of a common rule depending on the relative positions of the first vertices; defining a set of vertices as a known border, wherein the following steps are performed for selecting a symbol of the alphabet, for encoding the borders of the superpixels: a) determining the first vertex of the next superpixel border individuated by the common criterion; b) determining the next vertex to be encoded on the basis of the coding direction; c) selecting a first symbol (“0”) for encoding the next vertex if the next vertex of a border pertains to the known border, d) selecting a symbol (“1”; “2”) different from the first symbol (“0”) if the next vertex is not in the known border; e) repeating steps b), c), d) and e) until all vertices of the superpixel border have been encoded; f) adding each vertex of the superpixel border that was not in the known border to the set; g) determining the next superpixel whose border is to be encoded according to the common rule, if any; i) repeating steps a)-g) until the borders of all the superpixels of the image have being added to the known border.