Driver Facial Information Capture Using In-Vehicle Input Screens
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for inferring a vehicle driver's feelings by capturing their face image can be annoying for the driver, as they require displaying eye-guiding information to obtain accurate facial data.
Innovation Solution
An apparatus with a processor that captures facial information from a driver's face images, displays an input section on a vehicle's screen, and obtains correction information to determine the driver's looking direction and eye positions without annoying the driver, by strategically placing input sections on the screen and using classifiers to analyze eye and facial movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If display information is shown to guide the driver's eyes for obtaining face images, then measurement precision of facial information is improved, but driver annoyance increases and ease of operation deteriorates
Solution Approach 1:
The system uses the driver's natural interaction with the input section as both the operational interface and the facial data collection opportunity. The driver's own actions of looking at and interacting with the input section provide the facial information needed, eliminating the need for separate guidance displays.
Solution Approach 2:
The patent combines the function of obtaining facial information with the existing input operation. The same input section that receives driver interaction also serves as the target for capturing facial data, merging two functions into one unified interface.
2Measurement precision
If display information is shown to guide the driver's eyes, then measurement precision of looking direction is improved, but object-affected harmful factors increase due to driver annoyance
Solution Approach 1:
The driver's natural looking behavior at the input section during normal operation provides the looking direction data needed, without requiring additional guidance displays that would cause annoyance.
Solution Approach 2:
The patent converts the driver's necessary attention to the input section (which would otherwise be just a functional interaction) into a beneficial source of facial and looking direction data, turning a routine action into valuable measurement opportunity.
Data Source
AI summary
An apparatus for obtaining facial information obtains facial information indicating the state of the face of a driver of a vehicle from each of face images of the driver's face, causes a display mounted on the vehicle to show a screen including an input section for receiving input of information from the driver, and obtains correction information for correcting the facial information from a face image, of the face images, generated while the screen is displayed on the display.