Gaze State Classification Using Eye and Head Motion Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to accurately determine user intent based on eye movements, which are influenced by various factors such as task, state of mind, and body pose, leading to inefficiencies in user interaction with electronic devices.
Innovation Solution
A real-time gaze classification algorithm that classifies eye and head movements using video-based and retinal imaging, combined with scene understanding, to identify gaze shifting, holding, and loss events, independent of body pose and task, utilizing machine learning and physiological data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional eye tracking methods are used to determine user intent, then the system can detect eye movements, but the accuracy is insufficient due to variability in eye movement patterns caused by task, state of mind, and body pose
Solution Approach 1:
The patent segments eye movement analysis into distinct behavioral states (gaze shifting, gaze holding, loss events) with specific criteria for each state. This segmentation allows the system to accurately classify different types of eye movements based on quantitative metrics such as velocity thresholds and duration parameters, thereby improving measurement precision without requiring overly complex algorithms.
Solution Approach 2:
The patent employs multiple eye tracking parameters including pupil diameter, pupil location, gaze direction, and gaze angle simultaneously. By monitoring changes in these parameters over time and combining them with head pose data, the system achieves more accurate user intent detection. The use of velocity thresholds and duration parameters for differentiating gaze states represents a parameter-based approach to improving accuracy.
2Productivity
If real-time gaze classification is implemented, then user interaction responsiveness is improved, but computational latency may increase due to processing requirements
Solution Approach 1:
The patent implements preliminary classification by defining clear threshold-based criteria for different gaze behavior states before actual classification occurs. By pre-establishing velocity thresholds, duration requirements, and state transition rules, the system can rapidly classify eye movements in real-time without requiring complex post-processing computations, thus maintaining low latency while achieving accurate real-time responsiveness.
Solution Approach 2:
The patent processes only the necessary eye tracking parameters and head pose data required for gaze state classification, rather than analyzing all available sensor data. This selective processing approach maintains real-time performance by focusing computational resources on the most relevant parameters (pupil diameter, gaze direction, head pose) while ignoring extraneous information, thereby reducing overall processing latency.
3Measurement precision
If multiple eye tracking parameters are monitored simultaneously, then gaze behavior classification accuracy is improved, but data processing complexity increases
Solution Approach 1:
The patent merges eye tracking data with head pose data to achieve more accurate gaze behavior classification. By combining information from multiple sources (pupil diameter, pupil location, gaze direction, gaze angle, and head pose) into a unified classification framework, the system improves measurement precision while managing processing complexity through integrated analysis rather than separate processing of each parameter.
Solution Approach 2:
The patent introduces an intermediary classification layer that translates raw eye tracking parameters into meaningful gaze behavior states. This intermediary step processes the multiple parameters through defined thresholds and rules, converting complex multi-parameter data into simplified behavioral categories (gaze shifting, gaze holding, loss events), thereby reducing the complexity of subsequent processing while maintaining high classification accuracy.
Data Source
AI summary
Various implementations disclosed herein include devices, systems, and methods that determine a gaze behavior state to identify gaze shifting events, gaze holding events, and loss events of a user based on physiological data. For example, an example process may include obtaining eye data associated with a gaze during a first period of time (e.g., eye position and velocity, interpupillary distance, pupil diameters, etc.). The process may further include obtaining head data associated with the gaze during the first period of time (e.g., head position and velocity). The process may further include determining a first gaze behavior state during the first period of time to identify gaze shifting events, gaze holding events, and loss events (e.g., one or more gaze and head pose characteristics may be determined, aggregated, and used to classify the user's eye movement state using machine learning techniques).