Marker-Based Device Tracking Using Segmented Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing number of markers in captured images for tracking multiple devices in virtual reality environments leads to a significant increase in calculation complexity for estimating their positions and postures, making the process inefficient.
Innovation Solution
An information processing apparatus that acquires images of multiple devices with markers and estimates the position and posture information of one device first, then uses the remaining marker coordinates to derive the information for the other devices, reducing the overall calculation load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the information processing apparatus estimates position and posture information for multiple devices simultaneously using all marker images in captured images, then the tracking accuracy for all devices is maintained, but the calculation amount increases significantly
Solution Approach 1:
The patent segments the set of all marker images into multiple device-specific groups. By dividing the calculation task into separate estimation processes for each device, the system maintains accurate tracking for all devices while reducing the computational burden compared to processing all markers simultaneously. The segmentation allows independent optimization of each device's marker set.
Solution Approach 2:
The patent performs preliminary identification and classification of marker images into device-specific groups before the actual position and posture estimation. This preliminary action organizes the data structure in advance, allowing subsequent estimation processes to work with smaller, pre-sorted datasets, thereby reducing calculation complexity while maintaining accuracy.
2Adaptability or versatility
If the number of devices to be tracked increases, then the versatility of the virtual reality system is improved, but the calculation amount for estimating positions and postures increases according to the number of devices
Solution Approach 1:
The patent applies segmentation by dividing the total marker image set into multiple subsets corresponding to different devices. Each device's position and posture are estimated using only its specific marker subset, not all markers from all devices. This segmentation enables the system to track multiple devices efficiently without the calculation amount scaling linearly with the total number of devices.
Solution Approach 2:
The system performs preliminary classification of marker images into device-specific groups before estimation. This preliminary organization allows the system to handle multiple devices by processing pre-sorted marker sets, improving calculation efficiency while maintaining the ability to track any number of devices.
Data Source
AI summary
A captured image acquisition unit acquires an image obtained by capturing a first device that includes a plurality of markers and a second device that includes a plurality of markers. An estimation processing unit first derives position information and posture information of the first device on the basis of marker image coordinates in the captured image and then derives position information and posture information of the second device on the basis of remaining maker image coordinates excluding marker image coordinates of the first device.