Single-Channel Data Mapping for Multi-Channel Image Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video/image delivery systems face challenges in efficiently encoding and decoding additional data streams, such as depth data, vertex data, and index data, due to the lack of innate support for single-channel transport in existing multi-channel containers, leading to inefficient utilization of data capacity and the presence of compression artifacts.
Innovation Solution
A method and apparatus that convert single-channel data into n-dimensional values using a mapper, assigning these values to pixels in a virtual-image frame, and compressing the frame according to a container format, utilizing n-dimensional curves or tree partitions to ensure spatial and temporal coherence and inhibit compression artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If single-channel data is directly encoded into multi-channel containers, then the encoding process is simple, but the data capacity utilization is inefficient and compression artifacts appear
Solution Approach 1:
The patent applies dimensionality change by mapping single-channel scalar values into n-dimensional values using space-filling curves (Hilbert, Peano, Morton) or octree partitions. This transforms 1D data into 2D, 3D, or higher-dimensional representations that can be efficiently packed into multi-channel containers, improving data capacity utilization while maintaining spatial and temporal coherence of the original data.
2Device complexity
If single-channel data is directly encoded into multi-channel containers, then the encoding process is simple, but compression artifacts are present
Solution Approach 1:
By transforming single-channel data into n-dimensional representations using space-filling curves or octree partitions, the patent preserves spatial and temporal coherence relationships. This dimensional transformation allows the data to maintain its structural integrity during compression, reducing artifacts while remaining compatible with existing multi-channel container formats.
3Loss of information
If n-dimensional mapping is used to convert single-channel data, then data capacity utilization improves, but device complexity increases
Solution Approach 1:
The patent uses well-established space-filling curves (Hilbert, Peano, Morton) and octree partitioning algorithms to perform the dimensional transformation. These are deterministic mathematical methods that can be efficiently implemented in software, achieving good data capacity utilization without requiring complex hardware modifications to the existing video delivery pipeline.
Solution Approach 2:
The mapper acts as an intermediary component that sits between the single-channel data source and the multi-channel container encoding process. It performs the dimensional transformation without requiring modifications to the existing codec hardware, allowing the system to leverage existing infrastructure while improving data efficiency.
4Reliability
If n-dimensional mapping is used to convert single-channel data, then spatial and temporal coherence is preserved, but processing time increases
Solution Approach 1:
The patent performs the dimensional mapping operation during the encoding phase, transforming single-channel data into n-dimensional representations before compression. This preliminary action ensures that spatial and temporal coherence is preserved in the encoded data structure, allowing for efficient storage and transmission without requiring additional processing during playback.
Data Source
AI summary
Coding methods and apparatus for packing single-channel data into a multi-channel container, e.g., an MP4, TIFF, or JPEG container, to at least achieve good utilization of the container's data capacity. In some examples, a coding method comprises: converting a plurality of scalar values of a received data stream into a corresponding plurality of n-dimensional values, the converting being performed using a mapper; assigning each of the n-dimensional values as a pixel value to a respective pixel of a virtual-image frame, where n is an integer greater than one; and compressing the virtual-image frame according to a type of a container for image data. The mapper is configured to map a scalar value to a corresponding n-dimensional value based on a relationship represented by an n-dimensional curve or by a plurality of 2n-way tree partitions of n-dimensional space.