Handheld Object Pose Tracking Using Deep Search and Predictive Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for tracking the pose of handheld objects in augmented or virtual reality environments using optical tracking are computationally expensive due to the complexity of processing image data from multiple light sources.
Innovation Solution
A computing system that determines the pose of a handheld object with a plurality of light sources by acquiring image data, detecting light sources, and performing a search without previous pose data, using methods like Perspective-n-Point (PnP) and Kalman filters to fuse optical and inertial data, allowing for efficient pose determination and updating.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical tracking processes image data from multiple light sources to determine pose, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system performs a deep search without previous pose data to establish an initial pose determination. This preliminary action creates a baseline that enables subsequent predictive searches to leverage prior information, reducing computational complexity while maintaining precision
Solution Approach 2:
The system uses previously determined pose data to inform subsequent pose searches. By feeding back prior pose information into the tracking algorithm, the system reduces the search space and computational requirements for determining updated poses while maintaining accurate measurement
2Measurement precision
If deep search without previous pose data is performed, then measurement precision is improved, but loss of time increases
Solution Approach 1:
A deep search without previous pose data is performed as a preliminary action to establish an initial accurate pose. This one-time computational expense enables subsequent searches to benefit from prior information, reducing overall time loss while maintaining precision
Solution Approach 2:
The system alternates between deep searches (periodic recalibration) and predictive searches using previous pose data. This periodic approach ensures measurement precision is maintained through occasional full searches while reducing time loss through efficient predictive updates in between
Data Source
AI summary
Examples are disclosed herein that relate to determining a pose of a handheld object. One example provides a computing system configured to determine a pose of a handheld object comprising a plurality of light sources by acquiring image data of a surrounding environment, detecting a subset of light sources of the plurality of light sources of the handheld object in the image data, and performing a search, without using previous pose data, to determine the pose of the handheld object relative to the computing system. The computing system is further configured to use the pose determined to perform a later search for an updated pose of the handheld object, and if the later search fails to find the updated pose, determine the updated pose by again performing the search without using previous pose data.