Saccade Prediction for Imperceptible Redirected VR Walking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Navigating virtual environments larger than the available physical space in virtual reality is challenging due to unnatural locomotion techniques and the need for expensive hardware, which can disrupt the user's sense of presence and immersion.
Innovation Solution
A method using a machine learning model trained on eye-tracking data to predict saccades, allowing for real-time adjustments in the virtual environment without requiring additional hardware like eye-trackers, ensuring seamless navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If redirected walking techniques are used to navigate virtual environments larger than physical space, then users can traverse larger distances, but the manipulation of the virtual environment becomes perceptible to users, reducing their sense of presence
Solution Approach 1:
The system performs preliminary actions by predicting saccades before they occur using a machine learning model trained on eye-tracking data. By anticipating when the user's eyes will naturally shift, the system pre-prepares to apply environmental manipulations during these predicted saccade periods, ensuring the changes occur before the user's visual system can detect them.
Solution Approach 2:
The system exploits the brief period of visual suppression during saccades by rushing through environmental manipulations during these moments. The machine learning model identifies saccade timing, and the system applies rotations and warping effects quickly during these brief intervals when the user's visual system is less sensitive, thereby skipping over the detection window.
2Volume of moving object
If hardware-based redirected walking techniques are used, then users can navigate large virtual environments, but expensive additional hardware like eye-trackers or motion sensing systems are required
Solution Approach 1:
Instead of requiring expensive eye-tracking hardware, the system creates a computational copy or model of eye-tracking functionality using a machine learning model. This software-based copy replicates the essential function of detecting saccades by analyzing head movement patterns and predicting eye behavior, thereby achieving the same effect without the expensive physical hardware.
Solution Approach 2:
The system replaces the mechanical/optical eye-tracking system with a computational approach. Instead of using physical sensors to detect eye movements, the machine learning model substitutes by predicting saccades based on head movement data and physiological models of eye behavior, eliminating the need for complex hardware while achieving comparable functionality.
3Device complexity
If software-based redirected walking techniques are used, then cost is reduced, but visual artifacts and distortions appear in the virtual environment, negatively impacting user immersion
Solution Approach 1:
The system applies environmental manipulations in advance during predicted saccade periods. By preparing and executing rotations and warping effects during these brief moments when the user's visual system is suppressed, the manipulations occur before the user can perceive them, thereby eliminating visual artifacts from the user's experience.
Solution Approach 2:
The system converts the harmful effect of saccades (visual suppression) into a beneficial opportunity for environmental manipulation. Rather than treating saccades as interruptions that cause visual artifacts, the system exploits this natural visual suppression period to apply necessary environmental changes that would otherwise be perceptible, thereby turning a potential harm into a benefit for maintaining immersion.
Data Source
AI summary
Navigation of a virtual environment (VE) can mean navigating a VE that is spatially larger than the available Physical Tracked Space (PTS). Accordingly, the concept of redirected walking was introduced in order to provide a more natural way of navigating a VE, albeit with many restrictions on the shape and size of the physical and virtual spaces. However, prior art techniques have limitations such as requiring eye-tracking increasing costs and complexity of the user's head mounted display system or triggering major saccades in the user. Accordingly, the inventors have established a novel technique which overcomes these limitations. The technique is based on a machine learning/artificial intelligence system that predicts saccades such that the psychological phenomenon of inattentional blindness can be applied for re-directed walking without requiring the triggering major saccades in the users, complex expensive systems, etc.