Hierarchical Bytestream Decoding with Shared Parameters and Parallel Subsets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing encoding and decoding techniques for image and video data face challenges in optimizing metadata overhead and processing time, particularly in hierarchical encoding and parallel decoding of sparse data sets, leading to inefficiencies in data size reduction and memory utilization.
Innovation Solution
A method and apparatus for encoding and decoding data by dividing it into subsets, using attribute indicators to signal decoding parameters separately, allowing for grouping based on common features and optimizing metadata overhead while enabling parallel processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If data is divided into smaller subsets for enhanced entropy encoding, then compression effectiveness is improved, but metadata overhead increases
Solution Approach 1:
The patent merges multiple parameter sets into a single unified parameter set that can be applied across different data subsets. Instead of transmitting separate metadata for each subset, the system combines parameter information (such as entropy coding parameters, quantization parameters, and transformation parameters) into a consolidated metadata structure, thereby reducing overall metadata overhead while maintaining compression effectiveness.
Solution Approach 2:
The patent creates universal parameter sets that can serve multiple data subsets simultaneously. A single parameter set is designed to be applicable across different regions or types of data blocks, allowing the same metadata to control encoding for multiple subsets. This multi-functional approach eliminates the need for redundant parameter transmission and reduces metadata volume.
2Measurement precision
If sequential decoding is used to maintain accuracy, then decoding precision is improved, but processing speed deteriorates
Solution Approach 1:
The patent segments the encoded data stream into independent or semi-independent blocks that can be decoded in parallel. Each block contains self-contained information including all necessary parameters for its decoding, allowing multiple blocks to be processed simultaneously without compromising overall decoding accuracy. This segmentation enables parallel processing while maintaining the precision required for accurate reconstruction.
Solution Approach 2:
The patent performs preliminary organization of data blocks and parameters during the encoding phase, arranging them in a structure that facilitates parallel decoding. Metadata is prepared in advance to indicate independent decoding paths, and blocks are ordered or marked to allow the decoder to process multiple blocks concurrently while ensuring correct reconstruction order, thus improving processing speed without sacrificing accuracy.
Data Source
AI summary
There may be provided a method of decoding a received set of encoded data representing information that has been compressed, the method comprising: obtaining, a set of attribute indicators from the data set, each indicator of the set of indicators is associated with a subset of the data set; and, decoding a plurality of subsets of the data set, comprising: retrieving decoding parameters for each subset according to the attribute indicator associated with each subset; and, decoding each subset according to the retrieved decoding parameters for each subset.