Pose Prediction Models for AR Wearable Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Augmented reality wearable devices face challenges in providing real-time pose estimation due to latency in digital processing, leading to discomfort for users as there is a discrepancy between the displayed image and the actual pose of the device.
Innovation Solution
An electronic device that predicts future poses by combining multiple motion prediction models using a weighted average approach, incorporating image data and inertial data, and adjusts weights based on prediction time through machine learning to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pose estimation is performed through digital processing, then pose information can be obtained, but latency is generated causing discrepancy between displayed image and actual pose
Solution Approach 1:
The system performs preliminary pose prediction by combining multiple motion prediction models to forecast future pose before the actual pose estimation is completed. This predictive action compensates for the latency in digital processing, ensuring that the displayed image corresponds to the actual pose even though the processing is not yet complete.
2Measurement precision
If multiple motion prediction models are combined, then prediction accuracy is improved, but system complexity increases
Solution Approach 1:
The patent merges multiple motion prediction models (including physics-based models and machine learning models) into a unified framework that combines their strengths. By integrating these models and using a weighted average approach, the system achieves higher prediction accuracy while managing complexity through a structured combination methodology.
Solution Approach 2:
The system dynamically adjusts the weights of different prediction models based on prediction time and environmental conditions. This parameter change strategy allows the system to optimize accuracy by emphasizing certain models over others, effectively managing complexity through adaptive parameter adjustment rather than treating all models equally.
Data Source
AI summary
An electronic device includes a memory storing a plurality of motion prediction models, and a processor configured to estimate a current pose of the electronic device based on external data; predict a first motion of the electronic device at a future first prediction time through a first model in which the plurality of motion prediction models are combined for the future first prediction time, and calculate a first pose of the electronic device at the first prediction time based on the current pose and the first motion; and predict a second motion of the electronic device at a future second prediction time through a second model in which the plurality of motion prediction models are combined for the future second prediction time, and calculate a second pose of the electronic device at the second prediction time based on the first pose and the second motion.