Tile Boundary Distortion Reduction via Cross-Tile Wavelet Transform

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional remote desktop compression techniques suffer from tile boundary distortion due to edge-of-tile artifacts, limiting compression efficiency and impacting user experience.

Innovation Solution

The method involves processing tile coefficients to produce high and low frequency results, introducing an additional low frequency coefficient at tile boundaries by extrapolating from the last two coefficients of an adjacent tile, and applying discrete wavelet transforms to reduce distortion.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If each tile is individually compressed and decompressed using conventional tiling approach, then compression can be achieved, but edge-of-tile artifacts occur at tile boundaries causing distortion

Engineering Contradiction:
Improvecompression efficiencyVSAvoidimage quality at tile boundaries
Core Design Contradiction:
Loss of energyVSManufacturing precision

Solution Approach 1:

The patent merges adjacent tiles by combining their coefficient data and performing a cross-tile transform operation. This allows the transform to span tile boundaries, eliminating the artificial discontinuities that cause edge artifacts while maintaining compression efficiency through unified processing of boundary regions.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary padding of tile coefficients before the transform operation. By extending the coefficient arrays with zero values at the boundaries before transformation, the method prepares the data structure to handle boundary conditions smoothly, preventing artifact formation during the compressive transform process.

Inventive Principle:
Principle #10Preliminary action

2Quantity of substance

If conventional tiling compression is used, then data transmission size is reduced, but tile boundary distortion limits further compression

Engineering Contradiction:
Improvetransmitted data sizeVSAvoiddecoded image quality
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent merges adjacent tiles by combining their coefficient data and performing a cross-tile transform operation. This allows the transform to span tile boundaries, eliminating the artificial discontinuities that cause edge artifacts while maintaining compression efficiency through unified processing of boundary regions.

Inventive Principle:
Principle #5Merging (Combining)

3Productivity

If individual tile compression is applied, then processing can be done in parallel, but boundary artifacts reduce overall image quality

Engineering Contradiction:
Improvecompression processing speedVSAvoidboundary region quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent segments the image into tiles for parallel processing but applies a special cross-tile transform operation at the boundaries. This allows most of the image to be processed independently and in parallel, while only the boundary regions require coordinated processing, thus maintaining high productivity while improving boundary quality.

Inventive Principle:
Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach significantly reduces tile boundary distortion, enhancing bitmap image quality and allowing for improved compression ratios by ensuring only low-frequency coefficients are present at tile edges, thus improving the transmitted image quality.

Implementation Method 1

processing the coefficients to produce a high frequency result set and a low frequency result set, and introducing an additional low frequency tile coefficient to a boundary between two adjacent tiles by extrapolating a low frequency coefficient

Methodology Applied
Scientific EffectWavelet transform:

Data Source

PatentEP2951992B1Reducing tile boundary distortion
Publication Date: 2016.12.14 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP2951992B1 patent drawingFigure 1
  • EP2951992B1 patent drawingFigure 2
  • EP2951992B1 patent drawingFigure 3

AI summary

A method for removing boundary distortion includes receiving a one-dimensional input signal and determining whether the one-dimensional input signal includes an even number of data elements. If the one-dimensional input signal includes an even number of data elements, an extrapolation operation is performed on one of a first or second boundary of the one-dimensional input signal. The extrapolation operation produces one additional data element. The method may further include performing a mirroring operation on the signal data resulting from the extrapolation operation to produce a mirrored signal, and performing a transform operation the mirrored signal.