Gaze Tracking Target Positioning with Head Pose Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Gaze tracking systems have limited accuracy and resolution, making it difficult for users to navigate and select small or closely positioned content items using an eye-gaze interface, which can increase selection delays due to the need for additional steps and user input to enable display zoom modes.
Innovation Solution
A method that utilizes both gaze tracking data for coarse accuracy and head pose data for fine accuracy to position a target indicator within a display region, allowing users to refine the indicator's location with head movements, thereby improving selection precision and reducing delays by disabling the gaze tracking system after initial alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If gaze tracking data alone is used to position the target indicator, then the system is simple and fast, but the positioning accuracy is insufficient for small or closely positioned content items
Solution Approach 1:
The patent combines gaze tracking data and head pose data into a unified positioning system. The gaze tracking system provides coarse positioning while the head tracking system provides fine positioning, merging both data sources to achieve high-precision target indicator placement without requiring manual zoom operations.
Solution Approach 2:
The system introduces an intermediary processing layer that fuses gaze and head pose data. This intermediary mechanism reconciles the coarse gaze data with fine head pose data to generate precise target indicator positions, resolving the accuracy limitation of gaze tracking alone.
2Ease of operation
If display zoom modes are used to enlarge content items, then selection becomes easier, but additional user input steps and time delay are required
Solution Approach 1:
The patent replaces the mechanical zoom interface (requiring manual activation and deactivation) with a direct neural-controlled positioning system. Head movements naturally adjust the target indicator position without requiring separate zoom mode activation, eliminating the time delay associated with mode switching.
Solution Approach 2:
The system enables self-adjusting target indicator positioning through head movements. The interface automatically adapts to user intent by tracking head pose and adjusting the indicator position accordingly, without requiring the user to manually initiate zoom or positioning actions.
3Adaptability or versatility
If gaze tracking system remains active continuously, then positioning can be adjusted dynamically, but system resources are consumed
Solution Approach 1:
The system employs periodic activation of the gaze tracking system rather than continuous operation. Gaze tracking is activated during initial target acquisition and deactivated during fine positioning phases that rely on head pose data, reducing energy consumption while maintaining dynamic positioning capability when needed.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments that relate to positioning a target indicator via a display system are disclosed. For example, one disclosed embodiment provides a method for positioning a target indicator using gaze tracking data having a coarse accuracy from a gaze tracking system of a computing device. Head pose data having a fine accuracy greater than the coarse accuracy is received from a head tracking system. Using the gaze tracking data, an approximate user gaze region within a display region is determined, and the target indicator is displayed at an initial location within the approximate user gaze region. A reposition input from the user is received. In response, subsequently received head pose data is used to calculate an adjusted location for the target indicator. The target indicator is then displayed at the adjusted location.