Vehicle Object Recognition Using Ground-Plane Top-View Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing camera-based object detection systems lack depth information, making it difficult to accurately determine the position, speed, and heading of objects, and to track them effectively in autonomous vehicles.
Innovation Solution
An object recognition apparatus and method that uses a camera and a processor to determine a camera object box with projected points onto a ground plane, converts these points to a top-view perspective, and combines this with LIDAR data to improve accuracy in determining object length, width, and heading, enabling precise tracking and control of vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera-based object detection is used, then the system can identify objects in the vehicle's vicinity, but the accuracy of position information is insufficient due to lack of depth information
Solution Approach 1:
The patent projects 3D spatial coordinates onto a 2D ground plane, creating a top-view representation that preserves depth information through coordinate transformation. This dimensional transformation allows the system to maintain accurate position, speed, and heading information while working within the 2D image plane constraints of camera-based detection.
2Measurement precision
If LIDAR or radar is used to obtain position information, then depth accuracy is improved, but the system complexity and cost increase
Solution Approach 1:
The patent introduces a ground plane projection model as an intermediary that transforms camera image coordinates into top-view spatial coordinates. This mathematical transformation acts as a mediator, enabling the system to achieve LIDAR-like position accuracy using only camera data, thereby avoiding the need for additional depth-sensing hardware.
3Measurement precision
If multiple sensors (LIDAR, radar, camera) are combined to improve object detection accuracy, then measurement precision improves, but the device complexity and processing requirements increase
Solution Approach 1:
The patent makes the camera system multi-functional by enabling it to perform both 2D object detection and 3D position estimation through ground plane projection. This universal approach allows a single sensor type to fulfill multiple functions that traditionally required multiple different sensors, reducing system complexity while maintaining detection accuracy.
Data Source
AI summary
An object recognition apparatus may determine a first point where a portion of an object closest to a vehicle is projected onto a ground, a second point where a portion of the object furthest from the vehicle in a longitudinal direction is projected onto the ground, and a third point where a portion of the object furthest from the vehicle in a lateral direction is projected onto the ground, determine, based on a top-view perspective of the object, top-view points respectively corresponding to the first point, the second point, and the third point, determine, based on a second plurality of line segments connecting the top-view points, at least one of a length, a width, or a heading, track a position of the object based on at least one of the length, the width, or the heading of the object, and control the vehicle.