Layer Pre-Synthesis for Real-Time Multi-Layer Image Blending

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The high computational demand of real-time layer synthesis in multi-layer blending processes leads to frame freezing and poor user experience, particularly when a large number of layers are involved.

Innovation Solution

Pre-synthesizing foreground layers that meet associative or commutative conditions to reduce the number of layers participating in real-time synthesis, thereby improving efficiency and avoiding frame freezing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If real-time layer synthesis is performed with a large quantity of layers, then the layer synthesis result is complete and accurate, but the computing amount is huge causing frame freezing and poor user experience

Engineering Contradiction:
Improvelayer synthesis accuracyVSAvoidreal-time synthesis speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent segments the layer synthesis process into two distinct phases: a pre-synthesis phase where foreground layers are processed offline to generate intermediate images, and a real-time synthesis phase where only essential layers are combined. This segmentation divides the computationally intensive task into manageable parts, allowing complex foreground layer processing to occur separately from time-critical real-time rendering.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary synthesis of foreground layers before the actual real-time synthesis occurs. By pre-processing foreground layers to create intermediate images that capture essential visual information, the system prepares data in advance so that during real-time operation, less computational work is required, eliminating frame freezing while preserving synthesis quality.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If all foreground layers participate in real-time synthesis, then the synthesis result maintains high quality, but the quantity of layers increases computational load

Engineering Contradiction:
Improvesynthesis qualityVSAvoidcomputational complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent extracts foreground layers from the real-time synthesis pipeline and processes them separately through pre-synthesis. By taking out these computationally demanding layers and handling them in advance, the system reduces the number of layers that must be processed during real-time synthesis, thereby lowering computational complexity while maintaining synthesis quality through the use of generated intermediate images.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the processing parameters of foreground layers by applying pre-synthesis operations that transform them into intermediate images with optimized characteristics. This parameter transformation allows the foreground layer information to be represented in a form that is more efficient for real-time processing, reducing computational requirements while preserving visual fidelity.

Inventive Principle:
Principle #35Parameter changes

3Ease of operation

If layer synthesis is performed in real-time without pre-synthesis, then the processing is simpler, but frame freezing occurs due to high computing demands

Engineering Contradiction:
Improveprocessing simplicityVSAvoiddisplay stability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

By performing preliminary synthesis of foreground layers before real-time operation, the patent prepares necessary computational data in advance. This preliminary action ensures that during real-time synthesis, the system only needs to perform simpler operations with pre-processed data, maintaining display stability and preventing frame freezing while keeping real-time processing relatively simple.

Inventive Principle:
Principle #10Preliminary action

4Productivity

If the quantity of foreground layers is reduced through pre-synthesis, then real-time synthesis efficiency improves, but the pre-synthesis process adds initial processing time

Engineering Contradiction:
Improvereal-time synthesis efficiencyVSAvoidpre-synthesis processing time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent performs pre-synthesis operations in advance, before the user needs real-time synthesis. By moving the computationally intensive foreground layer processing to a preliminary phase that occurs before real-time operation, the system accepts initial processing time as a trade-off for achieving high real-time synthesis efficiency during actual use.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates intermediate images that serve as copies or representations of the synthesized foreground layer information. These intermediate images capture the essential visual data, allowing the system to work with simplified data structures during real-time operation rather than processing the original complex foreground layers repeatedly.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP4730101A1Layer synthesis method and electronic device
Publication Date: 2026.04.22 HUAWEI TECH CO LTD
  • EP4730101A1 patent drawingFigure 1(a)
  • EP4730101A1 patent drawingFigure 1(b)
  • EP4730101A1 patent drawingFigure 1(c)

AI summary

This application provides a layer synthesis method and an electronic device. In this application, foreground layers may be pre-synthesized, to effectively reduce a quantity of foreground layers participating in real-time synthesis. This improves real-time layer synthesis efficiency. The method includes: The electronic device obtains a background layer, an active layer, a first foreground layer, a second foreground layer, and a third foreground layer that are sequentially disposed. The electronic device synthesizes the first foreground layer, the second foreground layer, and the third foreground layer, to generate a first intermediate image. Then, the electronic device synthesizes the background layer, the active layer, and the first intermediate image, to generate a first target image.