Gaze Direction Tracking via Dual Eye Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for determining the direction of a person's gaze on a computing system face challenges in precision, mobility, calibration complexity, and real-time processing, leading to limitations in productivity and user freedom.
Innovation Solution
A method involving dual image processing of the eye, where the first processing determines the orientation based on static image analysis and the second processing provides kinematic information through successive image comparison, allowing for improved precision and reduced system requirements, enabling wireless and energy-efficient operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high precision gaze direction determination is implemented, then measurement precision improves, but device complexity and processing requirements increase
Solution Approach 1:
The patent segments the gaze determination process into two independent processing paths: (1) orientation processing that analyzes the aspect of eye zones to determine static eye orientation, and (2) kinematics processing that compares successive images to determine eye movement. This segmentation allows each path to use simpler, lower-resolution imaging while achieving high overall precision through combination of the two processing results.
Solution Approach 2:
The patent transitions from traditional single high-resolution 2D image processing to a multi-dimensional approach combining two separate lower-resolution processing streams (orientation and kinematics) that operate on different aspects of the same visual data. This dimensional transformation reduces the resolution requirements of individual sensors while maintaining or improving overall measurement precision.
2Measurement precision
If high resolution camera is used to acquire eye images, then measurement precision improves, but processing speed decreases and energy consumption increases
Solution Approach 1:
The patent divides the imaging task into two segmentation streams: orientation processing that uses lower resolution sufficient for detecting eye zone aspect changes, and kinematics processing that uses lower resolution sufficient for tracking movement between frames. This segmentation eliminates the need for a single high-resolution camera, thereby increasing processing speed while maintaining precision through the combined processing of both streams.
Solution Approach 2:
The patent changes the resolution parameter of the imaging system from high to low, compensating for the reduced individual image quality by introducing dual processing paths that extract different types of information (orientation from aspect, kinematics from temporal changes). This parameter change reduces computational load and increases processing speed while preserving measurement precision.
3Measurement precision
If complex calibration procedures are implemented, then initial precision improves, but calibration drift over time increases
Solution Approach 1:
The patent performs preliminary calibration to establish the relationship between eye orientation and gaze direction in the head frame of reference. This preliminary calibration, combined with the ongoing kinematics processing that tracks relative movements, maintains precision over time by continuously referencing back to the established calibration framework while accommodating natural drift through the dual-processing approach.
Solution Approach 2:
The patent implements feedback through the kinematics processing path that continuously monitors eye movement between successive images. This feedback mechanism detects and compensates for calibration drift by tracking actual eye behavior patterns, thereby maintaining long-term reliability and precision without requiring frequent recalibration.
4Measurement precision
If stationary remote camera is used to track user, then gaze direction can be determined, but user mobility is restricted and precision constraints are not met
Solution Approach 1:
Instead of using a stationary remote camera to track the user (external reference frame), the patent inverts the approach by embedding the imaging system on the user's device (mobile reference frame). This inversion allows the system to naturally adapt to user movement while maintaining precision through the dual processing of orientation and kinematics information in the device's coordinate system.
Solution Approach 2:
The patent transitions from a static stationary camera system to a dynamic mobile imaging system embedded on the user's device. This dynamic configuration automatically adapts to user mobility and head movements, maintaining measurement precision through real-time orientation and kinematics processing without restricting user freedom of movement.
Data Source
AI summary
The present invention relates to a method for determining the direction in which a user is looking, which includes acquiring images of the eye, in particular by means of an optical sensor, the method including: a) a first processing of the images, yielding information on the orientation of the eye according to the observation of an area of the eye in which the aspect varies with the rotation of the eye; and b) a second processing of the images, yielding information on the kinematics of the eye by comparing at least two consecutive images; method in which information is generated relating to the direction in which the user is looking relative to the head of the user, at least according to the information supplied by the first and second processes.