3D Eye Gaze Tracking for Accurate Driver Attention Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semi-autonomous driving systems fail to accurately determine driver attention, leading to potential accidents as they rely on user interactions, such as touching the steering wheel or facial expressions, which are not sufficient to ensure the driver's focus on the road.
Innovation Solution
A computing system that uses three-dimensional eye gaze vectors within the vehicle cabin to precisely determine where the driver is looking, analyzing the occupant's head and eye location, and applying machine learning models to differentiate between relevant in-vehicle and external objects, enabling more accurate assessment of driver engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional driver attention monitoring methods (steering wheel interaction, facial expressions) are used, then the system is simple to implement, but the measurement precision of driver attention is insufficient
Solution Approach 1:
The patent transitions from 2D facial expression analysis to 3D eye gaze vector analysis within the vehicle cabin space. By establishing a three-dimensional coordinate system and calculating gaze vectors with depth information, the system achieves more precise driver attention monitoring while maintaining reasonable system complexity through efficient computational geometry methods.
2Measurement precision
If 3D eye gaze vector analysis is implemented, then the measurement precision of driver attention improves, but the device complexity increases
Solution Approach 1:
The patent divides the vehicle cabin into multiple regions of interest (dashboard, windshield, side windows, rearview mirror) and determines which region the driver is looking at by intersecting the 3D gaze vector with these predefined spatial zones. This segmentation approach simplifies the complex problem of continuous 3D space analysis into discrete, manageable regions.
Solution Approach 2:
The patent introduces a coordinate transformation system that maps camera coordinates to vehicle cabin coordinates through an intermediary transformation matrix. This intermediary mathematical model enables accurate 3D gaze vector calculation without requiring complex direct measurements, bridging the gap between simple camera input and precise driver attention output.
3Reliability
If the system monitors driver attention continuously, then the reliability of safety detection improves, but the loss of time for processing increases
Solution Approach 1:
The patent pre-defines regions of interest and their boundary planes in the 3D coordinate system before runtime. During actual driver monitoring, the system only needs to calculate the gaze vector and determine which pre-defined region it intersects, rather than analyzing the entire continuous 3D space. This preliminary setup significantly reduces real-time computational burden while maintaining high reliability.
Data Source
AI summary
According to the techniques of this disclosure, a method includes capturing, using a camera system of a vehicle, at least one image of an occupant of the vehicle, determining, based on the at least one image of the occupant, a location of one or more eyes of the occupant within the vehicle, and determining, based on the at least one image of the occupant, an eye gaze vector. The method may also include determining, based on the eye gaze vector, the location of the one or more eyes of the occupant, and a vehicle data file of the vehicle, a region of interest from a plurality of regions of interests of the vehicle at which the occupant is looking, wherein the vehicle data file specifies respective locations of each of the plurality of regions of interest, and selectively performing, based on the region of interest, an action.