Eye Tracking Calibration via 3D Gaze Vector Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current eye tracking systems face challenges in accurately determining the gaze point of a user in 3D space, especially when multiple imaging sensors are involved, leading to inconsistencies and errors in tracking user attention.
Innovation Solution
The method involves obtaining a first 3D line passing through the center of the eye based on the relative orientation and position of the eye with respect to a first imaging sensor, and a second 3D line passing through the eye center. By determining the relative position of the eye center with respect to the imaging sensor using these lines, the system can accurately calibrate and correct gaze vector measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple imaging sensors are used to track eye gaze in 3D space, then the coverage and field of view are improved, but the measurement precision and reliability deteriorate due to inconsistencies and errors in tracking
Solution Approach 1:
The patent introduces a calibration target as an intermediary object that mediates between multiple imaging sensors and the eye tracking system. The calibration target provides known reference points that enable the system to calculate transformation matrices, thereby reconciling measurements from multiple sensors and improving overall measurement precision while maintaining extended field of view coverage
Solution Approach 2:
The patent changes the parameters of the imaging system by using multiple imaging sensors with different fields of view and combining their measurements through calibration. The calibration process adjusts transformation parameters (rotation and translation matrices) to harmonize data from multiple sensors, allowing the system to maintain high measurement precision across an expanded field of view
2Measurement precision
If a calibration target is introduced to improve measurement accuracy, then the measurement precision is improved, but the device complexity increases
Solution Approach 1:
The patent applies preliminary action by performing calibration before actual eye tracking measurements. The calibration target is positioned and measured in advance to establish transformation matrices that are then used during normal operation. This preliminary setup improves measurement precision without adding complexity to the ongoing tracking process, as the calibration data is stored and reused
Solution Approach 2:
The patent uses a virtual calibration target (displayed on a screen) instead of a physical calibration object. This copying approach allows the calibration process to be performed using the same imaging sensors and software infrastructure already present in the system, avoiding the need for additional specialized hardware and reducing overall device complexity while still achieving improved measurement precision
Data Source
AI summary
Disclosed herein are various methods of calibrating relative orientation between imaging sensor and spatial orientation sensor, and relative orientation between imaging sensors.