Bit-Depth Remapping for Guaranteed GPU Image Compression

Resolve Bottlenecks,
Find Innovative Solutions
Generate 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

VSEngineering 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

Engineering Contradiction:
Improvepower consumptionVSAvoidmemory bandwidth efficiency
Core Design Contradiction:
Use of energy by moving objectVSProductivity

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.

Inventive Principle:
Principle #10Preliminary action

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.

Inventive Principle:
Principle #35Parameter changes

2Quantity of substance

If lossy compression is used to meet compression thresholds, then storage space is reduced, but image quality deteriorates

Engineering Contradiction:
Improvestorage spaceVSAvoidimage quality
Core Design Contradiction:
Quantity of substanceVSLoss of information

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.

Inventive Principle:
Principle #3Local 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.

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If bit depth is reduced before compression, then compression ratio is improved, but data precision is lost

Engineering Contradiction:
Improvecompression ratioVSAvoiddata precision
Core Design Contradiction:
ProductivityVSMeasurement precision

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.

Inventive Principle:
Principle #10Preliminary action

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.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3588954A1Guaranteed data compression
Publication Date: 2020.01.01 IMAGINATION TECH LTD
  • EP3588954A1 patent drawingFigure 1
  • EP3588954A1 patent drawingFigure 2A~2C
  • EP3588954A1 patent drawingFigure 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.