Book Leaf Tracking via Fiduciary Marker and Edge Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Augmented reality systems face challenges in reliably tracking the position and orientation of fiduciary markers on book leaves, especially when they become distorted during page turning, leading to disruptions in image augmentation and user experience.
Innovation Solution
The implementation of a visibility constraint model and the use of fiduciary markers and non-alphanumeric patterns on book pages, combined with the Cell processor's ability to estimate page angles and orientations, allows for accurate tracking of turning book leaves by hypothesizing and evaluating edge positions and orientations, even when fiduciary markers are partially obscured.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fiduciary markers are used for tracking book leaves, then the system can provide reliable position and orientation estimation, but the markers become distorted and unreliable during page turning
Solution Approach 1:
The system divides the tracking task into two segments: fiduciary markers for reliable position/orientation estimation when visible, and edge detection for tracking during page turns when markers are obscured. This segmentation allows the system to use the appropriate method for each tracking scenario.
Solution Approach 2:
Edge detection acts as an intermediary method that bridges the gap when fiduciary markers become unreliable during page turning. The system uses edge detection to hypothesize leaf positions and orientations, providing continuous tracking coverage even when markers are distorted or hidden.
2Measurement precision
If the system uses traditional fiduciary marker detection, then it works well for stationary markers, but it fails when markers are partially obscured during leaf turning
Solution Approach 1:
The system dynamically switches between fiduciary marker detection and edge-based tracking methods based on visibility conditions. When markers become obscured during leaf turning, the system transitions to edge detection to maintain tracking continuity, and switches back to marker detection when markers become visible again.
Solution Approach 2:
The system uses feedback from visibility estimation to determine when to switch between tracking methods. By estimating whether fiduciary markers are visible or obscured, the system can adaptively select the appropriate tracking approach, ensuring continuous and accurate leaf position estimation.
3Reliability
If the system hypothesizes multiple edge positions to track turning leaves, then it can maintain tracking during page turns, but computational complexity increases
Solution Approach 1:
Instead of analyzing all possible edge positions, the system focuses on hypothesizing only the most likely leaf edge positions based on the visible fiduciary marker information. This partial action approach reduces computational complexity while maintaining tracking reliability by concentrating processing resources on the most probable tracking scenarios.
Data Source
Figure 1~2A
Figure 2B
Figure 3
AI summary
A method of book leaf tracking comprises receiving a video image comprising a book, estimating the current position and orientation of the book within the video image in response to a fiduciary marker of the book visible in the image, estimating the visibility of one or more predetermined features of the book, calculating a range of leaf turning angles that is consistent with the detected visibility of the or each predetermined feature of the book for the estimated current position and orientation of the book, and estimating the angle of a turning leaf of the book responsive to the calculated range.