3D Gaze Point Determination Using Scene Structure Intersection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to accurately determine the gaze endpoint at any three-dimensional object in 3D-space, particularly for determining the gaze endpoint at any arbitrary three-dimensional object in 3D-space, particularly for determining the gaze endpoint at any arbitrary three-dimensional object in 3D-space.
Innovation Solution
A system comprising an eye tracking unit, a head tracking unit, and a 3D scene structure representation unit is used to calculate the gaze endpoint by intersecting the gaze direction with a 3D structure representation, allowing determination of the object being gazed at in 3D-space.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If gaze direction is mapped to a predefined static plane, then the mapping process is simple, but the method cannot determine gaze endpoint on arbitrary 3D objects
Solution Approach 1:
The patent transitions from mapping gaze directions to a 2D static plane to intersecting gaze directions with a 3D scene structure representation. This dimensional extension from 2D to 3D space enables the system to determine gaze endpoints on arbitrary three-dimensional objects while maintaining mathematical tractability through structured 3D scene models.
Solution Approach 2:
The patent creates a virtual 3D scene structure representation that replicates the essential geometric features of the physical scene. This virtual model serves as a simplified copy that can be efficiently processed to determine gaze endpoints, avoiding the need for complex real-time processing of actual 3D object surfaces while preserving the necessary spatial relationships.
2Device complexity
If existing 2D plane mapping methods are used, then the system is computationally simple, but measurement precision of gaze endpoint on 3D objects is insufficient
Solution Approach 1:
By extending the mapping from 2D plane coordinates to 3D spatial coordinates, the patent achieves precise localization of gaze endpoints on three-dimensional objects. The 3D scene structure representation maintains computational efficiency through structured data organization while providing accurate depth information that 2D methods cannot capture.
Data Source
AI summary
A system for determining the gaze endpoint of a subject, the system comprising: a eye tracking unit adapted to determine the gaze direction of one or more eyes of the subject; a head tracking unit adapted to determine the position comprising location and orientation of the eye tracker with respect to a reference coordinate system; a 3D Structure representation unit, that uses the 3D structure and position of objects of the scene in the reference coordinate system to provide a 3D structure representation of the scene; based on the gaze direction, the eye tracker position and the 3D structure representation, calculating the gaze endpoint on an object of the 3D structure representation of the scene or determining the object itself.