Gaze Intent Detection via Segmented Localization and Triggering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current gaze interaction technologies face challenges in accurately determining user intent, particularly in distinguishing between localization and triggering operations, which can lead to errors in executing commands, especially in environments where precise control is necessary.
Innovation Solution
A user trigger intent determining apparatus that utilizes a gaze estimation model, including a neural network, to estimate the gaze location with high accuracy by preprocessing input images to identify key points and correct deviations, allowing for precise execution of commands based on user intent.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If gaze tracking data is used to determine user intent, then user interaction capability is improved, but accuracy in distinguishing localization from triggering operations deteriorates
Solution Approach 1:
The patent divides the gaze interaction process into two distinct phases: localization phase (detecting initial gaze location) and triggering phase (detecting confirmation gaze). By segmenting the interaction into separate detectable events with different characteristics, the system can accurately distinguish between merely looking at an element versus intending to activate it, resolving the contradiction between ease of operation and measurement precision.
Solution Approach 2:
The patent implements a preliminary localization detection before the final triggering decision. The system first detects the initial gaze location, then requires a separate confirmation gaze event to trigger the action. This preliminary action allows the system to gather more information about user intent before committing to an action, improving accuracy while maintaining ease of use.
2Device complexity
If simple gaze location detection is used, then device complexity is reduced, but reliability in command execution deteriorates
Solution Approach 1:
The patent segments the gaze detection process into multiple independent detection events (localization event and triggering event) rather than using a single complex decision algorithm. This segmentation maintains relative system simplicity while improving reliability by requiring distinct, detectable gaze behaviors for different operational phases.
Solution Approach 2:
The system uses feedback from sequential gaze detection events to improve command execution reliability. By monitoring the sequence and characteristics of gaze events (initial gaze, confirmation gaze, duration, movement patterns), the system can reliably distinguish between localization and triggering intent without requiring overly complex hardware or algorithms.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A user trigger intent determining method and apparatus is disclosed. The user trigger intent determining apparatus may obtain a first face image, obtain a second face image after a visual stimuli object is displayed, and determine a final gaze location by correcting a first gaze location estimated from the first face image based on a second gaze location estimated from the second face image.