3D Space Reconstruction from Photos Using Plane Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current three-dimensional space modeling methods are either time-consuming, require expensive equipment, or generate excessive data, making them impractical for widespread use, especially in applications requiring quick editing and accurate texture and dimension restoration.
Innovation Solution
A method involving photographing that reconstructs a three-dimensional space scene by importing photos, determining planes, marking spatial structures, and establishing a three-dimensional model using point coordinate information, allowing for easy editing and accurate two-dimensional floor plan generation, utilizing common capture devices like mobile phones and panoramic cameras.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If binocular stereo vision system is used, then three-dimensional geometric information can be obtained, but an extremely large number of photos need to be taken which is very time consuming
Solution Approach 1:
The patent segments the three-dimensional space into multiple planes (floor, ceiling, walls) and processes them separately. By dividing the space into discrete planar elements, the system can efficiently capture and reconstruct each plane independently, reducing the total number of photos needed compared to capturing the entire space uniformly with binocular stereo vision.
Solution Approach 2:
The patent transitions from three-dimensional binocular stereo vision to a two-dimensional plane-based reconstruction approach. By representing walls and surfaces as 2D planes with width, height, and position parameters, the system simplifies the reconstruction process and reduces computational complexity while maintaining accurate three-dimensional geometric information.
2Measurement precision
If laser point cloud technology is used, then spatial coordinates of each sampling point can be obtained, but it generates massive data which is not conducive to storage and processing
Solution Approach 1:
The patent extracts only the essential geometric parameters needed for three-dimensional reconstruction (plane equations, corner points, width, height, position) from the captured images. Instead of processing and storing massive point cloud data, the system extracts and stores only the critical plane parameters, significantly reducing data volume while preserving spatial coordinate information.
Solution Approach 2:
Instead of capturing dense point cloud data and then extracting plane information, the patent inverts the approach by directly capturing plane parameters (corners, edges, dimensions) from images and constructing three-dimensional models from these extracted features. This eliminates the need to handle massive intermediate point cloud data.
3Measurement precision
If binocular stereo vision system is used, then three-dimensional model can be reconstructed, but once the model goes wrong, the repair method is very complicated and cannot be operated by a non-professional person
Solution Approach 1:
The patent segments the three-dimensional model into independent editable planes (floor, ceiling, walls). Each plane can be individually selected, modified, and adjusted without affecting other parts of the model. This modular structure allows non-professional users to easily repair specific elements by simply editing the relevant plane parameters, eliminating the need for complex global model repairs.
Solution Approach 2:
The patent creates a dynamic, editable three-dimensional model where plane parameters (position, orientation, dimensions) can be freely adjusted after initial reconstruction. Users can interactively modify plane properties, add or remove planes, and reconfigure the model structure, transforming the rigid binocular stereo vision output into a flexible, user-editable format accessible to non-professionals.
4Measurement precision
If laser point cloud technology is used, then spatial distribution of target can be obtained, but it requires users to purchase point cloud equipment that is bulky, expensive and complicated to operate
Solution Approach 1:
The patent uses standard digital cameras or mobile phone cameras to capture images, which are then processed to extract plane parameters and reconstruct three-dimensional models. Instead of requiring specialized laser point cloud equipment, the system copies the functionality of expensive measurement devices using common imaging devices, making the technology accessible and eliminating the need for bulky, expensive hardware.
Solution Approach 2:
The patent replaces the mechanical laser scanning system with an optical imaging system followed by computational processing. Instead of using laser beams to physically measure spatial coordinates, the system uses optical images and mathematical algorithms to derive the same information, substituting a complex mechanical measurement system with a simpler optical-capture-and-compute approach.
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
Enables quick and convenient reconstruction of three-dimensional space models with accurate texture and dimensions, supporting a wide range of capture devices and avoiding incomplete scanning errors, while generating accurate two-dimensional floor plans.
Implementation Method 1
a camera is used to photograph a scene to be measured to obtain two-dimensional images of the scene
Implementation Method 2
obtain two-dimensional images of the scene through processing
Data Source
AI summary
A method for reconstructing a three-dimensional space scene based on photographing is disclosed, comprising the following steps: S1, importing photos of all spaces, and making the photos correspond to a three-dimensional space according to directions and viewing angles during capture, so that a viewing direction of each pixel, when viewed from the camera position of the three-dimensional space, is in line with that during capture; S2, regarding a room as a set of multiple planes, determining a first plane, and then determining all the planes one by one according to relationships and intersections between the planes; S3, marking a spatial structure of the room by a marking system and obtaining dimension information; and S4, establishing a three-dimensional space model of the room by point coordinate information collected in the step S3. In the present disclosure, the three-dimensional space model of a scene including dimensions and texture can be restored with no details lost; meanwhile, the three-dimensional space scene can be edited and modified quickly and conveniently, and a two-dimensional floor plan with dimensions and adequate accuracy can also be generated.