In-Situ Planar Natural Feature Targets for Single-Image 6DoF Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional augmented reality tracking systems require multiple reference images or user input to initialize a 3D map, leading to an unnatural user experience due to complex initialization procedures and limited tracking capabilities.
Innovation Solution
A method for initializing a three-dimensional target from a single reference image, allowing for real-time tracking in six degrees of freedom (6DoF) using a single camera without additional sensors, and refining the target with subsequent images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional tracking systems use multiple reference images or artificial markers to initialize 3D map, then tracking accuracy is improved, but device complexity and ease of operation deteriorate due to complex initialization procedures
Solution Approach 1:
The patent extracts and removes artificial markers from the tracking system, relying solely on natural features in the environment. The system initializes the 3D map using only naturally occurring visual features from captured images, eliminating the need for users to place or detect artificial markers, thereby simplifying operation while maintaining tracking accuracy through robust feature matching algorithms
Solution Approach 2:
The system performs preliminary initialization using a single reference image to establish the initial 3D map and camera pose before processing subsequent images. This preliminary action using natural features allows the tracking system to be ready for 6DoF tracking without requiring users to complete complex setup procedures with multiple reference images or artificial markers
2Manufacturing precision
If traditional methods require specific user motion sequences without visual feedback, then 3D reconstruction is achieved, but ease of operation and user experience deteriorate
Solution Approach 1:
The system provides continuous visual feedback by displaying the reconstructed 3D map and tracking status to the user in real-time. This feedback mechanism allows users to see the results of their camera movements immediately, eliminating the need to learn or remember specific motion sequences. The visual feedback confirms that 3D reconstruction is progressing correctly without requiring users to understand the underlying technical requirements
Solution Approach 2:
Instead of requiring users to perform specific motions to enable tracking, the system inverts the approach by enabling tracking from the first captured image with natural features and then using that initialization to guide subsequent processing. The system adapts to natural user movement rather than constraining users to predefined motion patterns, achieving 3D reconstruction through continuous image processing rather than prescribed sequences
3Device complexity
If a single camera is used for tracking, then device complexity is reduced, but measurement precision and reliability of tracking initialization deteriorate
Solution Approach 1:
The patent compensates for the limitations of a single camera by utilizing the temporal dimension through video sequences. The system captures multiple images over time and uses feature tracking across frames to initialize the 3D map and establish camera pose. This temporal dimension provides additional geometric constraints and redundancy that compensate for the lack of multiple simultaneous camera views, maintaining measurement precision while using only a single camera sensor
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed are a system, apparatus, and method for in-situ creation of planar natural feature targets. In one embodiment, a planar target is initialized from a single first reference image one or more subsequent images are processed. In one embodiment, the planar target is tracked in six degrees of freedom upon the processing of the one or more subsequent images and a second reference image is selected from the processed one or more subsequent images. In one embodiment, upon selecting the second reference image the planar target is refined to a more accurate planar target.