Incremental Pose Estimation for 3D Object Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for camera pose estimation in small-scale 3D object reconstruction, such as those using GPS and IMU sensors, are not suitable for handheld devices due to instability and error amplification, leading to incomplete reconstruction of 3D objects.
Innovation Solution
An electronic apparatus and method for incremental pose estimation and photographing, which includes capturing images of a 3D object at different positions using an image capturing unit and sensing unit, estimating a circumference for image capture, and adjusting a timing interval to trigger image capture based on displacements and angular displacements, allowing for re-estimation of central positions and circumferences to ensure complete image coverage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS and IMU sensors are used to assist feature matching process, then measurement capability is improved, but error amplification occurs and stability deteriorates in small-scale handheld measurement
Solution Approach 1:
The patent extracts and removes the unstable sensor data (GPS and IMU) from the measurement system for small-scale handheld applications. Instead of using these sensors that cause error amplification, the system relies solely on image-based feature matching and geometric constraints to achieve accurate 3D reconstruction without the harmful influence of unstable sensor data.
Solution Approach 2:
The patent replaces expensive and unreliable sensor systems with a simpler, more robust image-processing-based approach. By using only the camera and computational geometry methods, the system achieves measurement capability without relying on costly sensor hardware that proves unstable in handheld scenarios.
2Measurement precision
If all feature points are matched among all images, then reconstruction quality is improved, but computation resources are excessively required
Solution Approach 1:
The patent segments the feature matching process by establishing correspondence between sequential images rather than matching all feature points across all images simultaneously. This incremental approach divides the complex global matching problem into smaller, more efficient local matching tasks between adjacent frames, reducing overall computational burden while maintaining reconstruction quality.
Solution Approach 2:
The patent performs preliminary feature detection and description on individual images before the matching stage. By pre-processing and extracting key features in advance, the system reduces the computational complexity during the actual matching phase, as the expensive feature extraction is done once rather than repeatedly during matching operations.
3Ease of operation
If images are captured at fixed intervals, then capturing process is simplified, but incomplete coverage of 3D object occurs
Solution Approach 1:
The patent implements dynamic adjustment of the capture interval based on real-time analysis of image content and object characteristics. Rather than using a fixed time or distance interval, the system adaptively modifies the capture frequency to ensure complete coverage of the 3D object, capturing more images when complex geometries are detected and fewer images when surfaces are simple or already adequately covered.
Data Source
AI summary
An electronic apparatus and a method for incremental pose estimation and photographing are provided. In the method, at least two images of a 3D object are captured at different positions encircling the 3D object. Displacements and angular displacements are detected when capturing the images. Features of the 3D object in the images, the displacements and the angular displacements are used to estimate a central position of the 3D object and a distance between the electronic apparatus and the 3D object. A circumference suitable for capturing the images of the 3D object is estimated based on the distance and is divided into several segments, and a timing interval of a timer is adjusted based on a length of the segments. A camera of the electronic apparatus is triggered at intervals set by the timer to capture the images of the 3D object.