XR Object Tracking via Coordinate Alignment Without Wearables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing XR systems face high costs and user discomfort due to the use of wearable devices or optical sensors for motion capture.
Innovation Solution
A system and method utilizing a head-mounted display and image capture devices to align coordinate systems, determine vectors between objects, and update virtual objects in an extended reality scene, leveraging accelerometers, magnetometers, and machine learning for accurate motion capture without wearable devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If wearable devices with IMU are used for motion capture, then motion capture accuracy is improved, but device cost and user comfort deteriorate
Solution Approach 1:
The patent extracts the motion capture function from wearable devices and relocates it to the XR system's external cameras. The head-mounted display's cameras capture images of the user's body, and the controller processes these images to determine body pose and motion, eliminating the need for wearable IMU devices while maintaining motion capture accuracy.
Solution Approach 2:
The patent uses visual copying by capturing images of the user's body with cameras and processing these image data to reconstruct the user's motion. The controller extracts skeletal information and body pose from images, creating a digital representation of the user's movement without requiring physical sensors on the user's body.
2Measurement precision
If optical sensors or motion capture suits are used, then motion capture precision is improved, but user comfort deteriorates
Solution Approach 1:
The patent removes the requirement for optical sensors or motion capture suits by using the existing cameras on the head-mounted display. The system extracts motion information purely from visual images captured by these cameras, processing the images to determine body pose and skeletal information without any additional sensors on the user.
Solution Approach 2:
The head-mounted display's cameras serve multiple functions: providing the extended reality scene and simultaneously capturing images for motion capture. This multi-functionality eliminates the need for separate motion capture equipment, improving user comfort while maintaining precision through the same optical components.
3Measurement precision
If coordinate system alignment and transformation are performed, then object tracking accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent performs coordinate system alignment and transformation in advance by establishing the relationship between the head-mounted display's coordinate system and the image capture device's coordinate system before object tracking begins. The controller pre-processes the coordinate transformation matrices, so that during actual tracking, only simple vector transformations are needed rather than complex real-time coordinate alignments.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Provides a low-cost and comfortable solution for motion capture in XR environments by reducing the need for wearable devices and optical sensors, enhancing user experience.
Implementation Method 1
set a y-axis of the first coordinate system according to a first gravity direction measured by the first accelerometer
Implementation Method 2
set a z-axis of the first coordinate system according to a first geomagnetic direction measured by the first magnetometer
Data Source
AI summary
A system and a method of object tracking for an extended reality environment are provided. The method includes: providing, by a head-mounted display, an extended reality scene based on a first coordinate system; obtaining, by a first image capture device, a first image comprising a first object and a second object; aligning the first coordinate system with a second coordinate system to obtain a first transformation matrix; determining a first vector between the first object and the second object according to the first image; transforming the first vector to a second vector according to the first transformation matrix; and updating, by the head-mounted display, a virtual object in the extended reality scene according to the second vector, wherein the virtual object corresponds to the first object.