Monocular Camera 3D Position Estimation Using Reference Point Polygons
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in determining the three-dimensional location of objects within their environment, particularly from two-dimensional images captured by cameras, as the depth or distance to objects is not easily discernible, which can hinder obstacle avoidance and navigation.
Innovation Solution
The method involves determining the three-dimensional location of objects by first identifying reference points in the environment, projecting them into a two-dimensional image, selecting points that form a polygon around the object, and calculating the intersection point of a ray directed toward the object and this polygon to estimate the object's location in 3D space.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a monocular camera is used to capture images, then the device complexity is reduced, but the ability to determine 3D location and depth information deteriorates
Solution Approach 1:
The patent uses a monocular camera as an intermediary device to capture 2D images, then employs computational geometry methods (ray casting, polygon intersection) as mediators to infer 3D location information from the 2D image data, resolving the contradiction between simple device and precise measurement
Solution Approach 2:
The patent transforms the 2D image problem into a 3D space problem by projecting reference points into 3D coordinates and using ray casting techniques to determine depth, effectively adding a third dimension to the monocular vision system through computational methods
2Measurement precision
If multiple sensors are added to improve environment detection accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent makes the monocular camera perform multiple functions: capturing images for object detection, extracting reference points for 3D localization, and providing depth information through computational geometry, eliminating the need for separate sensors while maintaining multi-functional capability
3Measurement precision
If computational methods are used to determine 3D location from 2D images, then measurement precision improves, but processing time increases
Solution Approach 1:
The patent pre-establishes the relationship between 2D image coordinates and 3D world coordinates through reference points, and pre-defines the geometric models for ray casting and polygon intersection, allowing real-time 3D localization without complex runtime calculations
Data Source
AI summary
Disclosed herein are methods and systems for determining a location of an object within an environment. An example method may include determining a three-dimensional (3D) location of a plurality of reference points in an environment, receiving a two-dimensional (2D) image of a portion of the environment that contains an object, selecting certain reference points from the plurality of reference points that form a polygon when projected into the 2D image that contains at least a portion of the object, determining an intersection point of a ray directed toward the object and a 3D polygon formed by the selected reference points, and based on the intersection point of the ray directed toward the object and the 3D polygon formed by the selected reference points, determining a 3D location of the object in the environment.