Marker-Based Subject Tracking Under Occlusion and Marker Shift
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing subject tracking systems, particularly during medical procedures, face challenges with markers attached to the patient, such as skin markers that are prone to errors due to movement, occlusion, and false positives, which affect the robustness and accuracy of position and orientation tracking.
Innovation Solution
A device and method for subject tracking using a data input, processor, and output interface to identify and track multiple markers, ignoring temporarily displaced markers, and updating the reference model to maintain accurate tracking despite marker movement or occlusion, utilizing 2D or 3D image data from multiple directions to compensate for depth information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If skin markers are used for subject tracking, then the tracking system can provide six degree-of-freedom information on patient position and orientation, but the markers are prone to errors due to movement, detachment, or occlusion during the procedure
Solution Approach 1:
The system uses more markers than the minimum required for six degree-of-freedom tracking (at least four markers). By using an excessive number of markers, the system can tolerate the loss or displacement of some markers while maintaining accurate tracking through the remaining markers. The data processor identifies the largest matching subset of markers that matches the reference model, allowing the system to continue tracking even when some markers are temporarily ignored.
Solution Approach 2:
The system continuously compares the current marker positions with the reference model and provides feedback to the data processor. The data processor identifies which markers match the reference model and which do not, then adjusts the tracking calculation to use only the matching markers. This feedback mechanism allows the system to adapt to marker displacement or occlusion in real-time.
2Measurement precision
If multiple markers are used to ensure accurate tracking, then the system can maintain six degree-of-freedom information, but the complexity of identifying and matching all markers increases
Solution Approach 1:
The data processor extracts only the relevant information from the image data - specifically, the positions of identifiable markers. Rather than processing the entire image or considering all possible features, the system focuses exclusively on detecting marker positions and matching them to the reference model. This extraction approach simplifies the complexity by isolating the essential tracking information.
Solution Approach 2:
The system considers more markers than necessary (excessive action) but uses a systematic approach to match them. The data processor searches for the largest matching subset of markers that corresponds to the reference model, automatically identifying which markers are visible and which are occluded or displaced. This systematic matching process manages the complexity of multiple markers through algorithmic efficiency.
Data Source
Figure 1~2
Figure 3a~3c
Figure 4~5
AI summary
The present invention relates to tracking a subject, for example during medical examinations, interventions or other medical procedures. In order to provide improved subject tracking, a device (10) for subject tracking is provided that comprises a data input (12), a data processor (14) and an output interface (16). The data input is configured: to provide image data (18) of a subject comprising representation of at least four identifiable markers attached to the subject; and to provide a 3D reference model (20) of the markers which is based on an initial spatial arrangement of the markers attached to the subject. The data processor is configured: to identify the markers and their positions; to extract a marker arrangement based on the identified markers and their positions; to search for a spatial positioning of the 3D reference model matching with a maximum subset of markers of the extracted marker arrangement; to identify markers of the extracted marker arrangement that are not part of the matching subset as temporarily ignored; and to determine the spatial positioning of the 3D reference model that is matching with the maximum subset of the markers. The output interface is configured to provide the spatial orientation of the 3D reference model for a calculation of a spatial position of the subject.