Bit-plane coding for JPEG XS compression efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing bit-plane coding methods in JPEG XS are inefficient in terms of compression and coding complexity, particularly in handling transform coefficients with similar statistics across bit-planes.
Innovation Solution
Grouping coefficient groups into two sets and signaling whether all bit-planes within a group set are empty, with an option to indicate no prediction residual for certain group sets, allowing parallel determination of insignificance and enabling mode selection for improved coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If prediction residuals are coded for all coefficient groups, then coding accuracy is improved, but bitrate increases unnecessarily for insignificant coefficient groups
Solution Approach 1:
The patent extracts and separates the significance information from the prediction residual coding process. By introducing a separate significance flag that indicates whether a coefficient group is insignificant, the system can skip coding prediction residuals for those groups, thereby removing unnecessary bitrate consumption while preserving coding accuracy for significant groups.
Solution Approach 2:
Instead of applying prediction residual coding to all coefficient groups (excessive action), the patent applies it only to significant coefficient groups (partial action). The significance flag enables selective application of the coding process, avoiding waste on insignificant groups while maintaining accuracy where needed.
2Reliability
If bit-plane coding is performed on all transform coefficients, then coding completeness is improved, but coding complexity increases
Solution Approach 1:
The patent segments the transform coefficients into coefficient groups and further organizes them into group sets. By operating at the group set level with significance flags, the system reduces the granularity of processing from individual coefficients to groups, thereby reducing coding complexity while maintaining completeness through the organizational structure.
Solution Approach 2:
The patent merges multiple coefficient groups into group sets and applies a single significance flag to represent the entire group set. This merging reduces the number of individual coding operations required, simplifying the coding process while preserving all necessary information through the group set organization.
3Measurement precision
If significance information is coded for each coefficient group, then precision is improved, but bitrate consumption increases
Solution Approach 1:
The patent combines multiple coefficient groups into group sets and uses a single significance flag per group set rather than individual flags for each coefficient group. This merging reduces the total number of significance flags required, thereby reducing bitrate consumption while maintaining precision at the group set level.
Solution Approach 2:
The patent changes the dimension of significance representation from individual coefficient level to group set level. By operating at a higher dimensional level (group sets containing multiple coefficient groups), the system reduces the quantity of significance information that needs to be coded while preserving essential precision for reconstruction.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A coding efficiency improvement is achieved by performing bit-plane coding in a manner so that coefficient groups for which the set of coded bit-planes is predictively signaled in the datastream, are grouped in two group sets and if a signal is spent in the datastream which signals, for a group set, whether the set of coded bit-planes of all coefficient groups of the respective group set are empty, i.e. all coefficients within the respective group sets are insignificant. In accordance with another aspect, a coding efficiency improvement is achieved by providing bit-plane coding with group-set-wise insignificant signalization according to the first aspect as a coding option alternative relative to the signalization for group sets for which it is signaled that there is no coded prediction residual for the coded bit-planes for the claim groups within the respective group set.