Mediated Reality Input Mode Segmentation for Gaze Dependency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mediated reality systems, including augmented and virtual reality, face challenges in providing intuitive and efficient user interactions, particularly in transitioning between selection and action modes for virtual objects, often requiring continuous gaze detection which can be limiting and cumbersome.
Innovation Solution
A method that enables a first user-input mode for selecting virtual objects using a combination of gaze direction and selection gesture detection, followed by a second mode for performing actions on the selected object without relying on gaze detection, allowing for independent action gesture detection, and enabling selection of further virtual objects using gaze and gesture detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If continuous gaze detection is required for both selection and action operations, then selection accuracy is improved, but operation complexity and user fatigue increase
Solution Approach 1:
The interaction process is segmented into two distinct modes: selection mode requiring gaze + gesture for precise object selection, and action mode requiring only gesture for manipulating already-selected objects. This segmentation allows gaze detection to be disabled during action operations, reducing complexity and fatigue while preserving selection accuracy when needed.
Solution Approach 2:
The system dynamically switches between different input mode configurations based on the current operation state. During selection operations, the system enables both gaze and gesture detection for high precision. During action operations on selected objects, the system disables gaze detection and uses only gesture input, adapting the system complexity to the current task requirements.
2Measurement precision
If gaze detection is required for every action operation, then object selection precision is improved, but interaction efficiency deteriorates
Solution Approach 1:
The system performs object selection as a preliminary action that establishes the current operation target. Once an object is selected through the precise gaze+gesture mechanism, that selection state is maintained and reused for subsequent action operations. This eliminates the need to re-detect gaze for each action, improving efficiency while preserving selection precision when initially selecting objects.
Solution Approach 2:
The selected object state maintains continuity across multiple action operations. After initial selection using gaze detection, the system sustains the selected object reference through gesture-based actions without requiring repeated gaze detection. This continuous action model improves interaction efficiency by eliminating redundant detection cycles while preserving the precision established during the initial selection phase.
3Measurement precision
If gaze and gesture combination is used for all operations, then selection accuracy is improved, but ease of operation deteriorates
Solution Approach 1:
The operation sequence is divided into selection phase and manipulation phase. During the selection phase, the system requires both gaze and gesture input to ensure accurate object selection. During the manipulation phase, the system accepts only gesture input for actions on the already-selected object. This segmentation makes manipulation easier by removing the gaze requirement while preserving selection accuracy when it matters most.
Solution Approach 2:
The input requirements dynamically adapt based on the operation context. The system transitions from a dual-input mode (gaze + gesture) during selection to a single-input mode (gesture only) during manipulation. This dynamic adaptation optimizes ease of operation for the current task phase while maintaining selection accuracy during the critical selection moment.
Data Source
Figure 1A~1C
Figure 2A~3B
Figure 4~7B
AI summary
A method comprising: enabling a first user-input mode, in which user selection of a displayed virtual object in a virtual scene is in response to a combination of gaze direction detection and selection gesture detection; and enabling in response to user selection of the displayed virtual object in the virtual scene during the first user-input mode, a second different user-input mode for the selected virtual object, in which performance of an action in relation to the selected virtual object in the virtual scene is in response to action gesture detection without dependence upon gaze detection.