JPEG XS Coefficient Packing for Real-Time Encoding and Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The JPEG XS standard's encoding and decoding algorithms are inefficient for real-time processing of common video streams, such as those with a frame rate of 30 frames per second and a resolution of 1920x1080 pixels, due to sequential iteration over each bit of each coefficient.
Innovation Solution
A method for encoding and decoding that groups coefficients into buffers, accumulates bits, and interleaves arrays to form a packed data set, reducing the number of operations required.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If sequential iteration over each bit of each coefficient is used (as in JPEG XS standard), then encoding and decoding can be performed, but processing time is excessive for real-time video streams
Solution Approach 1:
The patent segments the sequence of coefficients into multiple groups, where each group contains a subset of coefficients. This segmentation allows parallel processing of different groups, reducing the overall processing time compared to sequential iteration through all coefficients one by one.
Solution Approach 2:
The patent performs preliminary actions by pre-calculating and storing bitplane information for each coefficient group before the main encoding/decoding process. This pre-processing enables faster retrieval and manipulation during actual compression/decompression operations.
2Productivity
If coefficients are processed individually in sequential order, then the algorithm is simple to implement, but it cannot achieve real-time processing performance
Solution Approach 1:
By dividing the coefficient sequence into groups and processing them in parallel, the patent achieves real-time performance while maintaining implementation simplicity through the use of standard buffer operations and bit manipulation techniques.
Solution Approach 2:
The patent introduces a new dimension of parallelism by processing multiple coefficient groups simultaneously rather than sequentially. This dimensional change from sequential to parallel processing enables real-time performance without significantly increasing algorithmic complexity.
3Productivity
If the standard JPEG XS packing method is used, then compatibility is maintained, but processing efficiency is insufficient for high-resolution video streams
Solution Approach 1:
The patent merges the processing of multiple coefficients into unified buffer operations. By accumulating bits from multiple coefficients into shared buffers and performing operations on entire groups rather than individual coefficients, the number of discrete operations is significantly reduced.
Solution Approach 2:
The patent creates universal buffer structures that can handle multiple coefficient groups with the same processing logic. This multi-functional approach allows the same code and data structures to process different groups in parallel, reducing overall operation count.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention relates to a method for encoding an input data set comprising a sequence of coefficients (10-43) into a packed data set (P1-P4), said method for encoding comprising a step of grouping the coefficients into groups (G1-G4) of coefficients, a step of accumulating a number of bits (100-434) of said groups into a set of buffers (B1-B4, B5-B8) so as to form an array of bits in each buffer, and a step of interleaving said arrays. The present invention also relates to an apparatus and a computer program for encoding an input data set comprising a sequence of coefficients into a packed data set, and to a method, apparatus, and computer program for decoding a packed data set obtainable by the method for encoding into an output data set.