Gesture-Based Authentication Using 3D Optical Sensor
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional secure access mechanisms, such as password logins, biometric logins, and token logins, face issues like forgetfulness, impersonation, and the need for additional devices, necessitating a more secure and user-friendly authentication method.
Innovation Solution
Implementing a gesture recognition system using an optical sensor and a processor to recognize unique three-dimensional limb movements, allowing users to create and authenticate with a sequence of gestures that are difficult to emulate, combining physical characteristics and movement patterns for secure login.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional passwords are used for security, then security is improved, but ease of use deteriorates due to complexity and forgetfulness
Solution Approach 1:
The patent replaces traditional password-based authentication (mechanical/memory system) with gesture-based authentication using an optical sensor and machine learning algorithm. The system captures video data of user gestures, extracts skeletal keypoints, and compares them against trained gesture patterns to authenticate users, eliminating the need for memorizing complex passwords while maintaining security.
2Ease of operation
If biometric login using fingerprint reader or facial recognition is used, then ease of use is improved, but security deteriorates due to impersonation risks
Solution Approach 1:
The patent moves from two-dimensional biometric data (fingerprint patterns, facial features) to three-dimensional spatial-temporal gesture data. The system captures the dynamic movement of body parts in 3D space over time, creating a more complex authentication space that is harder to replicate or impersonate while maintaining ease of use through natural gestures.
Solution Approach 2:
The patent uses dynamic gesture sequences instead of static biometric data. The gesture recognition system captures the temporal evolution of body movements, requiring the user to perform specific sequences of gestures in correct order and timing. This dynamic nature makes impersonation more difficult while keeping the authentication process simple and natural for the user.
3Reliability
If gesture recognition with detailed 3D limb movement analysis is implemented, then security is improved, but device complexity increases
Solution Approach 1:
The patent extracts only the essential features from complex gesture data for authentication. Instead of analyzing all video pixels or detailed motion vectors, the system identifies and tracks key skeletal joints (hand, wrist, elbow, shoulder, head, etc.) and their spatial-temporal relationships. This feature extraction approach maintains high security through detailed 3D movement analysis while reducing computational complexity by focusing on discriminative keypoints.
Solution Approach 2:
The patent creates simplified representations (copies) of complex gesture data for processing. The system generates skeletal keypoint sequences that capture the essential 3D movement patterns without requiring full video frame analysis. These extracted keypoint sequences serve as compressed representations that maintain authentication accuracy while reducing the complexity of subsequent processing and comparison operations.
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
Provides a secure and user-friendly login method that is difficult to replicate, ensuring the authorized user is authenticated while being easy to remember, thus enhancing security and accessibility.
Implementation Method 1
receiving, at the optical sensor, a sequence of user gesture input
Data Source
AI summary
An embodiment provides a method, including: activating, at an information handling device, an optical sensor; initiating, using a processor, an identification sequence; receiving, at the optical sensor, a sequence of user gesture input; determining if the sequence of user gesture input matches a predetermined sequence of user gesture input stored as training data, the determining comprising comparing, using the processor, data points derived from a three dimensional limb movement of the user gesture input with the stored training data; and providing an identification result. Other aspects are described and claimed.