Holographic Image Compression via Gabor Wavelet Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current technologies face challenges in achieving satisfactory compression rates for holographic images due to their large data size, which exceeds the capacity of current communication networks, especially when considering real-time transmission and display requirements.

Innovation Solution

A method that decomposes holographic images using a Gabor wavelet basis and selects relevant wavelet coefficients based on observer location, eliminating non-essential data by identifying cones of light intersecting with the observer's bounding volume, thereby reducing the data stream size to fit within network bandwidth.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If classical image compression techniques are applied to holographic sequences, then compression is attempted, but satisfactory results are not obtained because holographic images are diffraction patterns with poor correlation to the 3D scene

Engineering Contradiction:
Improvecompression rateVSAvoidreconstruction quality
Core Design Contradiction:
Loss of informationVSManufacturing precision

Solution Approach 1:

The patent transforms the holographic image from spatial domain to frequency domain using Gabor wavelet decomposition. This parameter transformation allows the compression algorithm to operate on wavelet coefficients rather than raw pixel data, achieving both high compression ratios and reconstruction quality by exploiting the energy compaction property of Gabor wavelets for holographic signals

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent segments the holographic image into multiple frequency bands through multi-level Gabor wavelet decomposition. Each band captures different spatial frequency characteristics, allowing selective compression where high-frequency components (less perceptually important) are compressed more aggressively while low-frequency components (more important for reconstruction) are preserved with higher fidelity

Inventive Principle:
Principle #1Segmentation

2Loss of information

If all wavelet coefficients are transmitted to ensure complete holographic information, then reconstruction quality is maintained, but data size exceeds network bandwidth capacity

Engineering Contradiction:
Improvereconstruction completenessVSAvoiddata size
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The patent applies local quality by differentiating transmission quality based on spatial location and observer position. Only wavelet coefficients corresponding to light cones that intersect with the observer's bounding volume are transmitted with high precision, while other coefficients are either compressed more aggressively or omitted entirely. This localized approach maintains reconstruction quality for the specific observer while dramatically reducing overall data transmission requirements

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent transmits only the necessary subset of wavelet coefficients required for a specific observer's view rather than all coefficients. By calculating which light cones from which wavelet locations intersect with the observer's bounding volume, the system transmits precisely the minimal set of coefficients needed, achieving partial action that optimizes bandwidth usage while maintaining sufficient reconstruction quality

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If the subset of wavelet coefficients is reduced to fit network bandwidth, then data transmission becomes feasible, but reconstruction quality may deteriorate

Engineering Contradiction:
Improvetransmission speedVSAvoidholographic reconstruction quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent incorporates feedback by using observer location information to dynamically determine which wavelet coefficients to transmit. The system calculates the bounding volume based on observer position and uses this feedback to select the appropriate subset of coefficients, ensuring that the transmitted data is always optimized for the current viewing conditions and maintains reconstruction quality

Inventive Principle:
Principle #23Feedback

4Loss of information

If Gabor wavelet decomposition is used to exploit directional spectrum properties, then relevant information extraction is improved, but computational complexity increases

Engineering Contradiction:
Improveinformation extraction efficiencyVSAvoidprocessing complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent performs Gabor wavelet decomposition as a preliminary action before compression and transmission. By pre-processing the holographic image into the Gabor wavelet domain, the system exploits the directional spectrum properties and energy compaction of Gabor wavelets for holographic signals, making subsequent compression more efficient and enabling the selective transmission approach to work effectively

Inventive Principle:
Principle #10Preliminary action

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 data size, enabling efficient transmission and display of holographic images by selecting only the relevant wavelet coefficients necessary for the observer's view, thus optimizing bandwidth usage.

Implementation Method 1

The invention consists in exploiting the directional nature of the spectrum associated with a Gabor wavelet and the relationship between the frequency localization of a Gabor wavelet and the direction of the light after diffraction through this wavelet at a point of the holographic image

Methodology Applied
Scientific EffectDiffraction: Diffraction

Data Source

PatentEP3097692B1Method for processing a sequence of holographic images, and associated computer program, devices, signals and devices.
Publication Date: 2019.11.06 ORANGE SA
  • EP3097692B1 patent drawingFigure 1~3b
  • EP3097692B1 patent drawingFigure 4~5
  • EP3097692B1 patent drawingFigure 6

AI summary

Method for processing a sequence of holographic images and associated computer program, devices and signal. The invention relates to a method for processing a sequence of holographic images with a view to its reconstruction by a holographic display device and delivery to at least one observer, characterised in that, each holographic image being decomposed in a Gabor wavelet basis into a set of wavelet coefficients, the method comprises the following steps, implemented for each image of the sequence: obtaining information representative of the decomposition in the Gabor wavelet basis; obtaining information representative of a location of said at least one observer in a frame of reference of the display device; and selecting a subset of wavelet coefficients depending on the obtained wavelet and location information.