3D Model Generation System Using Panorama Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing techniques for estimating the size of an object from photographed images require capturing images from multiple directions, which is time-consuming and increases processing load on computers.

Innovation Solution

A 3D model generating system that splits a panorama image into multiple images, generates partial 3D models for each image, and arranges them in a virtual space to create an overall 3D model of a room, reducing the need for multiple captures and processing load.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple photographed images from different directions are used to estimate object size, then measurement precision is improved, but loss of time increases and productivity decreases

Engineering Contradiction:
Improveobject size estimation accuracyVSAvoidphotographing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the panorama image into multiple sub-images, each capturing a different direction. By processing these segmented images independently to generate partial 3D models, the system achieves accurate object size estimation from a single comprehensive panorama shot, eliminating the need for time-consuming multiple captures from different positions.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If multiple photographed images from different directions are used to estimate object size, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improveobject size estimation accuracyVSAvoidprocessing load
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the complex task of processing multiple full images into simpler sub-tasks by segmenting the panorama into directional sub-images. Each sub-image is processed independently to create partial 3D models, which are then integrated. This segmentation reduces the computational complexity compared to processing multiple complete photographs.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from 2D image processing to 3D model generation by creating partial 3D models from segmented panorama images. This dimensional transformation allows the system to leverage spatial relationships in three dimensions, improving measurement precision while managing processing complexity through structured 3D reconstruction.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Manufacturing precision

If multiple photographed images are used to generate 3D model, then manufacturing precision of the model is improved, but productivity decreases

Engineering Contradiction:
Improve3D model accuracyVSAvoidmodel generation speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent performs preliminary segmentation of the panorama image into directional sub-images before 3D model generation. By pre-processing the image in this structured manner, the system prepares data that can be efficiently converted into accurate partial 3D models, achieving high manufacturing precision without the time penalty of multiple separate photographing sessions.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3467788B1Three-dimensional model generation system, three-dimensional model generation method, and program
Publication Date: 2022.08.03 RAKUTEN GROUP INC
  • EP3467788B1 patent drawingFigure 1
  • EP3467788B1 patent drawingFigure 2
  • EP3467788B1 patent drawingFigure 3

AI summary

Processing load on a computer can be reduced while saving the labor of photographing an object to estimate a size of the object. Photographed image acquiring means (101) of a 3D model generating system (1) acquires a photographed image in which a first object and a second object are disposed in a real space. Virtual image generating means (105) generates a virtual image in which a first model representing the first object that is a standard size and a second model representing the second object that is not the standard size are disposed in a virtual space. Parameter determining means determines a size parameter indicating a size of the second object based on a result of comparing the photographed image with the virtual image.