Line-of-Sight Detection Using Inter-Frame Gaze Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing line-of-sight detection technologies suffer from high power consumption and poor real-time performance, particularly in applications like VR games, due to the need for continuous frame-by-frame image analysis.
Innovation Solution
A method that compares inter-frame variation amounts between current and reference eye image frames to determine a target gaze direction, using pre-generated gaze directions when minimal change is detected, thereby reducing computational overhead and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If continuous frame-by-frame image analysis is performed for line-of-sight detection, then real-time detection capability is maintained, but power consumption increases significantly
Solution Approach 1:
The patent implements periodic action by performing full frame-by-frame image analysis only when inter-frame variation exceeds a threshold, otherwise using pre-generated gaze directions. This periodic intensive analysis combined with periodic light analysis resolves the contradiction between maintaining real-time detection and reducing power consumption.
Solution Approach 2:
The patent applies preliminary action by pre-generating gaze directions for reference image frames before they are needed for comparison. This allows the system to quickly determine whether full analysis is necessary by comparing against pre-computed values, reducing the frequency of power-intensive operations while maintaining real-time responsiveness.
2Measurement precision
If full frame-by-frame image analysis is performed, then detection accuracy is maintained, but computational load increases
Solution Approach 1:
The patent implements partial action by performing complete image analysis only when necessary (when inter-frame variation exceeds threshold), and using partial analysis (comparing against pre-generated directions) during stable periods. This selective application of analysis depth maintains accuracy when needed while reducing computational load during stable gaze conditions.
Solution Approach 2:
The system uses feedback by continuously monitoring inter-frame variation and dynamically adjusting the analysis strategy based on detected changes. When variation exceeds thresholds, full analysis is triggered; otherwise, lighter comparison operations are used, creating a feedback-driven adaptive computational load that maintains accuracy while reducing overall complexity.
3Productivity
If inter-frame comparison is performed frequently, then real-time performance is improved, but power consumption increases
Solution Approach 1:
The patent implements periodic action by performing intensive inter-frame comparison operations only when necessary to detect genuine gaze changes, rather than continuously. This periodic intensive comparison maintains real-time performance for detecting meaningful events while reducing power consumption during stable periods.
Solution Approach 2:
The patent uses copying by creating and storing pre-generated gaze directions for reference frames, then copying/comparing against these pre-computed values instead of performing full analysis on every frame. This copying approach maintains real-time detection capability while significantly reducing the computational and power costs of frequent comparisons.
Data Source
AI summary
Embodiments of the disclosure provides a method, an apparatus of line-of-sight detection, an electronic device and a storage medium. The method includes: obtaining a current image frame and a first reference image frame, where in the current image frame is an eye image collected in real time, and the first reference image frame is an eye image collected before the current image frame; obtaining an inter-frame variation amount of the current image frame relative to the first reference image frame; and deriving a target gaze direction based on the inter-frame variation amount, where the target gaze direction is a first gaze direction or a second gaze direction, the first gaze direction is a gaze direction pre-generated based on the first reference image frame, and the second gaze direction is a gaze direction generated in real time based on the current image frame.