Bit-Depth Remapping for Guaranteed GPU Image Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data compression methods for image data in graphics processing units (GPUs) face challenges in reducing memory bandwidth and storage space efficiently, particularly in mobile devices, due to varying compression ratios and the need for random access, which can lead to increased power consumption and memory inefficiencies.
Innovation Solution
A method is introduced that determines the bit depth of input data by comparing its total size to pre-defined threshold values, using a mapping parameter to reduce bit depth and encode it for data compression, ensuring a guaranteed compression threshold is met, thereby reducing memory bandwidth and storage space while maintaining random access capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If data compression is applied to reduce memory bandwidth and storage space, then power consumption is reduced, but compression ratio variability and random access requirements can lead to increased power consumption and memory inefficiencies
Solution Approach 1:
The patent performs preliminary analysis of the data block to determine the actual compression ratio before final compression. This allows the system to pre-calculate the required bit depth reduction and configure the compression parameters in advance, avoiding repeated compression attempts and associated power consumption.
Solution Approach 2:
The patent dynamically adjusts compression parameters (bit depth reduction levels) based on the analyzed data characteristics and required compression ratio. By changing parameters adaptively rather than using fixed compression settings, the system optimizes the balance between compression efficiency and power consumption for different data types.
2Quantity of substance
If lossy compression is used to meet compression thresholds, then storage space is reduced, but image quality deteriorates
Solution Approach 1:
The patent applies different bit depth reduction levels to different channels or subsets of channels based on their individual characteristics and importance. This allows selective preservation of quality in critical channels while achieving higher compression in less critical channels, optimizing the overall balance between storage space and image quality.
Solution Approach 2:
The patent performs partial bit depth reduction rather than uniform full reduction across all data. By selectively applying compression to specific portions of the data based on analysis results, the system achieves the required compression threshold while minimizing quality loss in essential data regions.
3Productivity
If bit depth is reduced before compression, then compression ratio is improved, but data precision is lost
Solution Approach 1:
The patent performs preliminary analysis to determine the optimal bit depth reduction level before actual compression. This pre-calculation ensures that bit depth is reduced only to the extent necessary to meet compression targets, preserving maximum precision while achieving required compression ratios.
Solution Approach 2:
The patent dynamically adjusts the bit depth reduction parameter based on data characteristics and compression requirements. Rather than applying fixed bit depth reduction, the system adapts the reduction level to maintain adequate precision while achieving target compression ratios for different data types and scenarios.
Data Source
Figure 1
Figure 2A~2C
Figure 3A~3B
AI summary
A method of data compression is described in which the total size of the compressed data is determined and based on that determination, the bit depth of the input data may be reduced before the data is compressed. The bit depth that is used may be determined by comparing the calculated total size to one or more pre-defined threshold values to generate a mapping parameter. The mapping parameter is then input to a remapping element that is arranged to perform the conversion of the input data and then output the converted data to a data compression element. The value of the mapping parameter may be encoded into the compressed data so that it can be extracted and used when subsequently decompressing the data.