Eye Tracking Illuminator Control via Predicted Glint Position
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing eye tracking systems face challenges in accurately tracking gaze information due to the obstruction of glints on the pupil edge and increased processing latency caused by multiple illuminators, which can degrade accuracy and complexity.
Innovation Solution
The system selectively powers on and off illuminators based on predicted glint positions relative to the pupil edge, using two-dimensional image analysis without requiring a three-dimensional cornea center position, to optimize power consumption and reduce processing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If multiple illuminators are powered on to illuminate the eye for imaging, then the eye tracking system can capture sufficient light for image generation, but glints may fall on the pupil edge and obstruct gaze information detection
Solution Approach 1:
The system predicts the position of glints in advance based on current eye position and glint motion trends. By determining whether glints will fall on the pupil edge in the next image before actually capturing it, the system can proactively adjust illuminator states to prevent glint obstruction, rather than reacting after the fact.
Solution Approach 2:
The system dynamically adjusts the power state of illuminators based on real-time eye tracking data and predicted glint positions. Instead of keeping all illuminators statically on or off, the system switches illuminators on or off depending on the predicted need, adapting to changing eye positions and glint trajectories to optimize both illumination and accuracy.
2Illumination intensity
If multiple illuminators are powered on to ensure adequate illumination, then image capture is sufficient, but processing complexity and latency increase
Solution Approach 1:
Instead of keeping all illuminators on continuously, the system applies partial action by selectively powering on only the subset of illuminators that are predicted to be needed for the next image capture. This reduces the number of glints and processing requirements while maintaining sufficient illumination for accurate eye tracking.
Solution Approach 2:
The system extracts and removes unnecessary illuminators from the active set by predicting which illuminators will not be needed based on eye position and glint trajectory analysis. By taking out redundant illuminators before image capture, the system reduces processing complexity and latency while maintaining adequate illumination.
3Illumination intensity
If all illuminators remain powered on continuously, then adequate illumination is maintained, but power consumption increases
Solution Approach 1:
The system employs periodic action by switching illuminators on and off in a rhythmic pattern based on predicted eye movement and glint position. Instead of continuous operation, illuminators are activated only during periods when they are predicted to be needed for accurate eye tracking, reducing overall power consumption while maintaining illumination continuity when required.
Solution Approach 2:
The system performs preliminary action by predicting future illumination needs based on current eye position and glint trajectory. By determining in advance which illuminators will be needed for the next image capture, the system can power them on just in time and power off others, avoiding continuous power consumption while ensuring illumination is available when needed.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach improves the accuracy of eye tracking by minimizing the number of active illuminators, reducing processing complexity, and conserving power, while maintaining accurate gaze detection.
Implementation Method 1
images of an eye may be generated and processed to detect the pupil and glints
Implementation Method 2
a light source, such as one including a light emitting diode (LED), is powered on to illuminate the eye and for the imaging
Data Source
AI summary
Techniques for controlling light sources used in eye tracking are described. In an example, an eye tracking system generates a first image and a second image showing at least a portion of the user eye illuminated by a predetermined set of illuminators of the eye tracking system. The eye tracking system determines a first position of a glint in the first image and a second position of the glint in the second image. Each of the first position and the second position is relative to a pupil edge. The eye tracking system predicts a third position of the glint relative to the pupil edge based on the first position and the second position. Further, the eye tracking system determines, from the predetermined set, an illuminator that corresponds to the glint and determines, based on the third position, whether to power off the illuminator to generate a third image of at least the portion of the user eye.