AR-Based 3D Building Reconstruction With Pose Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3-D building models face challenges in generating accurate and realistic renderings due to costly computing requirements for large image inputs and inadequate information from sparse data, while augmented reality (AR) devices suffer from sensor drift and noise, leading to location inaccuracies without effective integration with other image data for precise measurements.
Innovation Solution
A method leveraging augmented reality frameworks to capture images of building structures using non-camera anchors, identify reference poses, and incorporate real-world poses to generate accurate 3-D models by selecting candidate poses based on ratio comparisons and applying a scaling factor, which includes illumination data for realistic rendering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large image inputs (such as video input) are used for 3-D model reconstruction, then the accuracy and completeness of the 3-D model is improved, but the computing cost and resource requirements increase significantly
Solution Approach 1:
The patent extracts and utilizes illumination data from augmented reality world maps, separating this specific useful information from the complete AR dataset. This extraction allows the system to leverage lighting information for realistic rendering without processing the entire large-scale image input, thereby reducing computing costs while maintaining model accuracy.
Solution Approach 2:
The patent employs augmented reality frameworks to simultaneously achieve multiple functions: capturing spatial information for 3-D reconstruction, obtaining illumination data for realistic rendering, and gathering environmental context. This multi-functionality eliminates the need for separate data collection processes, reducing overall computing resources required.
2Use of energy by moving object
If image sets with sparse data are used for 3-D model reconstruction, then the computing cost is reduced, but the information captured is inadequate for realistic rendering and accurate measurements
Solution Approach 1:
The patent introduces augmented reality world maps as an intermediary data source that bridges the gap between sparse image inputs and comprehensive scene understanding. The AR framework processes environmental information and provides structured data including illumination, spatial relationships, and semantic context, enabling accurate rendering and measurements from limited input images.
Solution Approach 2:
The system performs preliminary processing of the environment using augmented reality frameworks before the actual 3-D reconstruction. By pre-extracting illumination data and spatial information from AR world maps, the system prepares essential data structures in advance, allowing sparse image inputs to be efficiently integrated without losing critical information.
3Ease of manufacture
If augmented reality device data is used alone for 3-D reconstruction, then the process is simplified and can be performed with standard devices, but sensor drift and noise cause location inaccuracies
Solution Approach 1:
The patent merges augmented reality device data with traditional photogrammetry techniques and external reference data. By combining multiple data sources including AR world maps, image features, and geometric constraints, the system compensates for sensor drift and noise in AR devices, achieving accurate location and measurement data while maintaining accessibility to standard devices.
4Measurement precision
If traditional photogrammetry methods are used without augmented reality integration, then measurement accuracy can be achieved, but the process requires specialized equipment and complex workflows
Solution Approach 1:
The patent enables the system to self-correct AR device inaccuracies by automatically integrating multiple data sources and using algorithmic fusion of photogrammetry and AR data. The system performs self-calibration and error compensation without requiring specialized equipment or manual intervention, maintaining measurement accuracy while simplifying the workflow to use standard devices.
Data Source
Figure 1A
Figure 1B
Figure 1C~1D
AI summary
System and method are provided for scaling a 3-D representation of a building structure. The method includes obtaining images of the building structure, including non-camera anchors. The method also includes identifying reference poses for images based on the non-camera anchors. The method also includes obtaining world map data including real-world poses for the images. The method also includes selecting candidate poses from the real-world poses based on corresponding reference poses. The method also includes calculating a scaling factor for a 3-D representation of the building structure based on correlating the reference poses with the selected candidate poses. Some implementations use structure from motion techniques or LiDAR, in addition to augmented reality frameworks, for scaling the 3-D representations of the building structure. In some implementations, the world map data includes environmental data, such as illumination data, and the method includes generating or displaying the 3-D representation.