3D Vehicle Surround Visualization Using Gaussian Sphere Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional top view output apparatuses for visualizing a vehicle's surroundings suffer from distortion of ground objects, limited visibility, and poor image quality, especially in wider areas, due to their inability to account for the continuity of ground plane objects and insufficient information from remote cameras, making them impractical for real-world applications like reverse parking.
Innovation Solution
A method and apparatus for synthesizing 3D images using wide-angle cameras, which correct lens distortion and map images onto a Gaussian sphere, allowing for accurate representation of ground and non-ground plane objects in a 3D virtual space, enhancing visibility and reducing distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If top view output apparatuses use conventional algorithms including lens distortion correction and homography conversion, then ground plane information can be continuously represented, but ground objects other than ground plane cannot be guaranteed continuous and their distortion is considerably increased
Solution Approach 1:
The patent applies spherical projection to map images from multiple cameras onto a spherical surface, preserving the curved nature of ground objects rather than forcing them into a planar representation. This curvature-based approach maintains the continuous structure of ground objects like curbs and sidewalks while avoiding the severe distortion that occurs when spherical surfaces are flattened into 2D plane images.
Solution Approach 2:
The patent transitions from 2D planar image synthesis to 3D spherical image synthesis by projecting camera images onto a spherical surface. This dimensional change allows ground objects to maintain their spatial relationships and continuity in three-dimensional space, avoiding the distortion inherent in two-dimensional planar representations.
2Area of stationary object
If top view output apparatuses represent a wider area beyond 2m boundary from vehicle, then visual area is expanded, but image quality degradation and distortion become serious
Solution Approach 1:
By using spherical projection, the system can represent a 360-degree panoramic view around the vehicle while maintaining image quality across the entire field of view. The spherical surface naturally accommodates wide-angle camera views without the severe distortion that occurs in planar projections at the edges of wide fields of view, enabling high-quality representation of areas beyond 2m from the vehicle.
3Device complexity
If top view output apparatuses make information about surroundings planar, then processing is simplified, but distortion of ground objects becomes easily found and visualization becomes impractical
Solution Approach 1:
The patent maintains spherical geometry throughout the image synthesis process, from projection to display. This approach preserves the natural curvature of ground objects and eliminates the distortion artifacts that occur when spherical scenes are flattened. The spherical representation provides intuitive and accurate visualization of the vehicle surroundings while maintaining reasonable processing complexity through established spherical projection algorithms.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution provides improved visibility and accuracy of objects around a vehicle by representing them in a 3D format with low distortion, overcoming limitations of conventional systems and enabling more stable operation and practical use, especially in scenarios like reverse parking.
Implementation Method 1
a sphere mapping unit configured to obtain optical parameters for the correction of lens distortion including an optical center of a lens using the photographed images, and to map each of the images to a surface of a Gaussian sphere using obtained optical center
Data Source
AI summary
The present invention relates to three-dimensional visualization of the surrounding images of a vehicle, and comprises the steps of: enabling a plurality of wide angle cameras provided at a vehicle to receive a plurality of photographed images for reference patterns formed on the ground; extracting feature points from the photographed reference patterns and estimating a relative location and an installation angle of each camera using the known physical location information of the extracted feature points; obtaining optical parameters comprising an optical center of a lens for correcting lens distortion using the photographed images and mapping each image on a surface of a Gaussian sphere using the obtained optical center; changing an angle and distance such that the Gaussian sphere and the real reference patterns correspond to each other using the relative location and the installation angle of the estimated camera and arranging the images in a three-dimensional virtual space; and obtaining a three-dimensional single image by mapping each image arranged in the three-dimensional virtual space to an inner surface of the three-dimensional sphere corresponding to one large sphere.