Driver View Overlay on Exterior Video for Action-Cause Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for displaying vehicle exterior videos fail to accurately represent the driver's actual field of view and the actions leading to that view, such as dozing or using a smartphone, making it difficult to determine the cause of accidents or near-misses.
Innovation Solution
A method that combines vehicle interior and exterior videos using AI to analyze feature points like the driver's face and hands, superimposing a frame of view and action display on the exterior video to accurately depict the driver's field of view and actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If only vehicle exterior video is displayed, then the video content is simple and easy to process, but the viewer cannot grasp the driver's actual field of view or actions
Solution Approach 1:
The patent combines vehicle exterior video with driver action analysis results into a single composite video output. The controller overlays action information (such as driver gestures, eye movements, and facial expressions) onto the exterior video, allowing viewers to simultaneously see both the road environment and the driver's actions without requiring separate displays or complex post-processing.
Solution Approach 2:
The controller acts as an intermediary that processes raw video data from the in-vehicle camera and integrates it with exterior video. It performs action analysis on driver images, determines action types based on recognized features, and synthesizes this information into an enhanced video output that bridges the gap between simple exterior video and complex driver monitoring data.
2Measurement precision
If motion analysis of driver features is performed, then the accuracy of driver action recognition is improved, but the processing time and computational load increase
Solution Approach 1:
The patent segments the driver image analysis into distinct feature detection components (eye movements, facial expressions, hand gestures) that can be processed independently. The controller identifies specific action types by analyzing individual features separately and then integrates these results, allowing for optimized processing of each feature type and reducing overall computational complexity.
Solution Approach 2:
The system performs action analysis selectively rather than continuously analyzing all video frames. The controller determines when action analysis is necessary based on detected changes or specific conditions, and processes only the relevant portions of the video stream, reducing overall processing time while maintaining recognition accuracy for critical driver actions.
Data Source
AI summary
A generation method by a controller, the generation method includes: acquiring a vehicle interior video from an in-vehicle camera and a vehicle exterior video from an out-vehicle camera; executing a motion analysis of feature points of at least one of a face, an eye, or a hand of a driver based on the vehicle interior video; and generating video data displaying a frame of view of the driver and motion information of the driver superimposed on the vehicle exterior video based on the motion analysis.