Eye Gaze Selection Using Time Window Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing eye gaze tracking systems for selecting user interface objects often result in inaccurate determinations due to synchronization issues between eye gaze location and selection input, especially when users move quickly between objects or experience system latencies.
Innovation Solution
Evaluating a plurality of eye gaze samples within a time window around the selection input to determine the intended user interface object, considering stability metrics and contextual information to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single eye gaze sample at the time of selection input is used to determine UI object selection, then the system response is fast and simple, but the selection accuracy deteriorates due to synchronization issues and system latencies
Solution Approach 1:
The system collects and stores multiple eye gaze samples before the selection input is received, preparing the data in advance. When a selection input occurs, the system already has a buffer of gaze data to evaluate, rather than waiting to collect samples after the input is received. This preliminary data collection resolves the contradiction by having accuracy-critical data ready without adding complex real-time evaluation mechanisms.
Solution Approach 2:
The patent transitions from evaluating a single eye gaze sample (one-dimensional point in time) to evaluating multiple eye gaze samples across a time window (adding the time dimension). This dimensional expansion allows the system to assess gaze stability and intent over time, improving selection accuracy without requiring complex spatial adjustments.
2Measurement precision
If multiple eye gaze samples within a time window are evaluated to improve selection accuracy, then the accuracy improves, but the processing time and computational load increase
Solution Approach 1:
The system evaluates only a subset of collected eye gaze samples within the time window rather than processing every single sample. By selecting representative samples or using sampling thresholds, the system achieves sufficient accuracy without the computational overhead of evaluating all available data points, thus reducing processing time while maintaining improved selection accuracy.
Solution Approach 2:
Eye gaze samples are collected and pre-processed in advance within the time window before the selection input occurs. This preliminary collection and initial filtering of gaze data reduces the computational burden at the moment of selection, allowing the system to evaluate multiple samples accurately without excessive processing delays when the actual selection decision is made.
3Loss of information
If eye gaze samples are collected continuously over time, then the data availability for accurate selection improves, but the data synchronization difficulty increases
Solution Approach 1:
The system introduces a time window as an intermediary framework that bounds and structures the collection of eye gaze samples. This time window acts as a mediator between continuous gaze data flow and discrete selection events, organizing samples into manageable, synchronized intervals that reduce the complexity of handling continuous data while maintaining comprehensive gaze information for accurate selection.
Data Source
AI summary
Various embodiments relating to selection of a user interface object displayed on a graphical user interface based on eye gaze are disclosed. In one embodiment, a selection input may be received. A plurality of eye gaze samples at different times within a time window may be evaluated. The time window may be selected based on a time at which the selection input is detected. A user interface object may be selected based on the plurality of eye gaze samples.