Gaze Tracking Unlocking Resists Shoulder-Surfing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional unlocking methods, such as passwords, patterns, fingerprint, and facial recognition, are vulnerable to attacks like shoulder-surfing and presentation attacks, and are not user-friendly for individuals with disabilities.
Innovation Solution
An eye gaze tracking system that randomly selects images with predefined feature points, requiring users to follow a specific sequence of gaze points to unlock devices or applications, providing a touch-free, robust, and secure method resistant to various attacks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If password or pattern unlocking methods are used, then the device can be unlocked, but the unlocking process is observable by unauthorized individuals enabling shoulder-surfing attacks
Solution Approach 1:
The patent replaces visible mechanical input methods (typing passwords, drawing patterns on screens) with an optical/biological measurement method (eye gaze tracking). The system captures eye movement data through cameras and processes it to determine unlocking actions, making the unlocking process invisible to observers while maintaining user accessibility.
2Ease of operation
If face recognition unlocking is used, then the device can be unlocked, but it becomes vulnerable to presentation attack techniques
Solution Approach 1:
The patent changes the measurement parameter from static facial features (which can be replicated in photos or masks) to dynamic eye movement trajectories. By capturing the temporal sequence and pattern of eye movements across multiple images, the system creates a unique behavioral signature that is difficult to replicate through presentation attacks.
3Ease of operation
If fixed trajectory eye gaze unlocking is used, then the unlocking process is simple, but it is vulnerable to prediction and manipulation attacks
Solution Approach 1:
The patent transforms the unlocking mechanism from a static fixed trajectory to a dynamic adaptive system. The system randomly selects from multiple images and dynamically adjusts the expected gaze trajectory based on the selected image content, making prediction attacks ineffective while maintaining ease of use through natural eye movement patterns.
4Ease of operation
If traditional unlocking methods are used, then the device can be unlocked, but they are not friendly to disabled people
Solution Approach 1:
The patent substitutes mechanical interaction requirements (hand dexterity for typing or drawing) with ocular movement, which is accessible to users with various physical disabilities. This allows users with motor impairments, arthritis, or other disabilities that affect hand coordination to unlock devices through natural eye movements.
Data Source
AI summary
A method and apparatus for unlocking a device or an application on the device based on gaze tracking. An image randomly selected from a set of images. Each image in the set of images includes a plurality of feature points and a sequence of feature points is pre-defined for each image. The selected image is presented to a user. A trajectory of gaze points of the user on the selected image presented to the user is determined. It is then determined whether the trajectory of gaze points of the user matches the pre-defined sequence of feature points on the selected image. The device or the application on the device is unlocked if the trajectory of gaze points of the user matches the pre-defined sequence of feature points on the selected image.


