Head Pose Attention Tracking via Distance Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in determining a person's direction of attention, especially when the person is not within close proximity to the camera or when multiple individuals are present, as they struggle to accurately track gaze direction due to limitations in eye tracking technology, particularly at distances greater than a meter or with multiple viewers.
Innovation Solution
The system employs head pose determination using a camera, leveraging depth information from dual sensor cameras or infrared/laser rangefinders to estimate the distance and orientation of a person's head, correlating this with possible objects of attention on a viewing plane, allowing for the identification of a cone of view without relying solely on eye direction, and switches to gaze tracking when closer for precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If eye tracking is used to determine direction of attention, then measurement precision is improved for close proximity, but reliability deteriorates at distances greater than a meter
Solution Approach 1:
The system dynamically switches between head pose determination (for distant users) and eye gaze tracking (for close users) based on the detected distance from the user to the camera. This dynamic adaptation allows the system to maintain reliable attention direction determination across varying distances by selecting the appropriate tracking method for each distance range.
Solution Approach 2:
The patent introduces an intermediate approach using head pose determination as a mediator when eye tracking becomes unreliable at distances greater than a meter. By using head orientation as an intermediate indicator of attention direction, the system extends reliable tracking capability to distant users while maintaining the precision of eye tracking for close users.
2Measurement precision
If eye tracking is used to determine direction of attention, then measurement precision is improved, but device complexity increases due to multiple sensor requirements
Solution Approach 1:
The system implements a universal attention tracking framework that can operate with different methods (head pose determination or eye gaze tracking) depending on the situation. The dual sensor camera system serves multiple functions: it can perform head pose determination using standard imaging, and it can perform eye gaze tracking when conditions are favorable, making the system adaptable to various user distances and scenarios without requiring completely separate hardware systems.
3Reliability
If head pose determination is used for distant users, then reliability is improved, but measurement precision deteriorates compared to eye tracking
Solution Approach 1:
The system dynamically selects the appropriate tracking method based on user distance. For distant users where head pose determination provides sufficient reliability, it uses that method. For close users where eye tracking can provide higher precision, it switches to eye gaze tracking. This dynamic selection optimizes the balance between reliability and precision based on the specific operational context.
Data Source
AI summary
An object that has attracted a viewer's interest is determined using head pose and distance information. In one example a viewer's head is detected at a camera of a computing system. A distance from the head to the camera is determined. A pose of the head is determined as a yaw angle of the head. A location is determined using the distance and the yaw angle, and an object of attention is determined using the location.