Neural Mosaic Image Generation Without Source Image Libraries

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods require a base image and source images to construct a mosaic image, limiting their applicability when neither is available.

Innovation Solution

A computing apparatus and method that generates a mosaic image using a neural network to create source images from an input image without relying on a base image or source images, incorporating features like color, texture, and geometric information, and evaluates esthetics for natural and user-desired composition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional mosaic construction methods are used requiring base image and source images, then the mosaic construction process is simple and straightforward, but the applicability is limited when neither base image nor source images are available

Engineering Contradiction:
ImproveapplicabilityVSAvoidprocess complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system enables self-service by allowing the base image to generate source images automatically through neural network processing. The base image is divided into multiple regions, and source images are synthesized from these regions without requiring external source image inputs, making the system adaptable to scenarios where no pre-existing source images are available.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

A neural network serves as an intermediary component between the base image and the generated source images. The neural network processes the base image regions and transforms them into suitable source images, enabling the mosaic construction process to function without direct human intervention or pre-prepared source images.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If source images are generated from base image regions using neural network, then the applicability without source images is improved, but the computational complexity and processing time increase

Engineering Contradiction:
Improvecapability to generate without source imagesVSAvoidcomputational energy
Core Design Contradiction:
Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The base image is segmented into multiple smaller regions, and the neural network processes each region independently to generate corresponding source images. This segmentation approach reduces the computational burden compared to processing the entire base image at once, while still enabling source image generation without pre-existing source images.

Inventive Principle:
Principle #1Segmentation

3Ease of operation

If manual selection and arrangement of source images is used, then the composition control is precise, but the time consumption and labor requirements increase

Engineering Contradiction:
Improveoperation convenienceVSAvoidprocessing time
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system performs self-service by automatically generating and arranging source images based on the base image regions. The neural network synthesizes source images and the system automatically positions them in the mosaic grid, eliminating the need for manual selection and arrangement operations, thus reducing both time consumption and operational complexity.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3809364B1Computing apparatus and method for constructing a mosaic image
Publication Date: 2026.04.22 SAMSUNG ELECTRONICS CO LTD
  • EP3809364B1 patent drawingFigure 1
  • EP3809364B1 patent drawingFigure 2
  • EP3809364B1 patent drawingFigure 3

AI summary

Provided are a computing apparatus for constructing a mosaic image and an operation method of the same. The computing apparatus includes: a memory storing one or more instructions; and a processor configured to execute the one or more instructions stored in the memory to: segment an input image into a plurality of sub areas to obtain a plurality of sub area images, extract a feature from each of the plurality of sub area images, generate a plurality of source images respectively corresponding to the plurality of sub areas using an image generation neural network, the image generation neural network using, as a condition, the feature extracted from each of the plurality of sub area images, and combine the plurality of source images respectively corresponding to the plurality of sub areas to generate a mosaic image.a