Mixed Reality Graphical Environment Tuning with Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Understanding and predicting user behavior in mixed reality (MR) spaces is complex due to the immersive nature and wide range of user interactions, making it difficult to capture and interpret relevant data effectively.
Innovation Solution
A machine-learning toolkit is employed to analyze user behavior data, extracting patterns and optimizing MR spaces for user engagement and productivity by training a machine learning model on usage scenarios from multiple training instances and adjusting features based on real-time user interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning model is trained on multiple training instances with usage data, then the accuracy of usage scenario detection is improved, but the complexity of the system increases
Solution Approach 1:
The system performs preliminary training of the machine learning model using multiple training instances and usage data before actual operation. This pre-computation phase establishes the model's accuracy in advance, allowing the operational system to benefit from high detection accuracy without bearing the computational burden of continuous retraining, thus resolving the contradiction between accuracy and operational complexity.
Solution Approach 2:
The patent introduces an intermediary machine learning model that bridges raw usage data and usage scenario detection. The model acts as a mediator that has already learned patterns from training instances, enabling accurate scenario detection during operation without requiring the operational system to directly process and analyze raw usage data, thereby reducing operational complexity while maintaining high accuracy.
2Productivity
If MR environment features are dynamically adjusted based on real-time usage data, then user engagement is improved, but the processing time and computational resources increase
Solution Approach 1:
The system implements a feedback mechanism where usage data from users is continuously monitored and fed into the trained machine learning model to detect usage scenarios in real-time. The detected scenarios then trigger automatic adjustments to MR environment features, creating a closed-loop system that enhances user engagement through responsive adaptation without requiring manual intervention, thus improving productivity while managing processing time through automated decision-making.
Solution Approach 2:
The patent dynamically changes parameters of the MR environment (such as layout, content, or interaction modes) based on usage scenario detection results. By pre-training the machine learning model to recognize patterns and make decisions, the system can rapidly adjust environmental parameters in response to detected scenarios, enhancing user engagement while minimizing processing time through efficient parameter optimization rather than exhaustive analysis.
Data Source
AI summary
Techniques are described for training a machine learning model on parameters calculated from usage parameters of a plurality of training instances of a mixed reality graphical environment (MRGE) to determine usage scenarios using a supervisory signal and then using the trained machine learning model to ascertain usage scenarios for non-training instances of the MRGE to determine usage scenarios. The ascertained usage scenarios may then be used to dynamically adjust features of the non-training instances of an MRGE.