Panoramic Image Generation via 3D Point Model Projection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating panoramic images from multiple optical cameras suffer from parallax errors when images are not acquired from the same position and orientation, especially when using moving vehicles, limiting the field of view and resolution, and are not suitable for increasing the number of cameras.
Innovation Solution
A method involving the creation of a 3D point model with position and color information, where images are projected onto this model to generate a panoramic image from a virtual viewpoint with minimal parallax errors, using a Light Detection And Ranging (LIDAR) unit to scan the area and combine images from various positions, allowing for a 360-degree field of view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple optical cameras are used to increase resolution and field of view, then image resolution and field of view are improved, but parallax errors occur when images are not acquired from the same position
Solution Approach 1:
The patent introduces a virtual viewpoint as an intermediary concept that mediates between multiple physical camera positions. By projecting all images onto a common 3D point model from this virtual viewpoint, the system eliminates parallax errors while preserving the benefits of multiple cameras for increased resolution and field of view.
Solution Approach 2:
The patent transitions from 2D image plane geometry to 3D spatial geometry by creating a point model with three-dimensional coordinates. This dimensional change allows images from different positions to be integrated without parallax by using depth information to correctly position features from each image in the unified 3D space.
2Productivity
If images are acquired from a moving vehicle to cover large areas, then productivity and coverage area are improved, but parallax errors increase due to changing camera positions
Solution Approach 1:
The patent performs preliminary actions by first creating a complete 3D point model of the environment before generating the panoramic image. This pre-established spatial framework allows subsequent image projections from moving vehicle positions to be accurately integrated without parallax errors, enabling high productivity while maintaining precision.
Solution Approach 2:
The patent changes the parameter reference frame from camera-centric 2D coordinates to environment-centric 3D coordinates. By transforming all image data into the unified 3D point model coordinate system, the system can accommodate varying camera positions during vehicle movement while eliminating parallax errors in the final panoramic output.
3Manufacturing precision
If cameras are placed one behind the other on a vehicle to reduce parallax, then parallax errors are reduced, but the number of cameras that can be used is limited
Solution Approach 1:
Instead of arranging cameras to physically align their viewpoints (traditional approach), the patent inverts the problem by using a virtual viewpoint that does not correspond to any physical camera position. This allows cameras to be distributed anywhere on the vehicle while all images are projected from the same virtual location, eliminating parallax and enabling unlimited camera placement flexibility.
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
This method enables the generation of panoramic images with high resolution and minimal parallax errors, suitable for large areas, by accurately positioning and orienting images from multiple sources, providing detailed information about infrastructure and environments.
Implementation Method 1
using a Light Detection And Ranging (LIDAR) unit to scan the area
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
A method for generating a panoramic image comprising the steps of: providing a 3D point model of an area surrounding a virtual viewpoint, acquiring multiple images of the area surrounding the virtual viewpoint, projecting the acquired images onto the 3D point model, and generating the panoramic image using the thus obtained 3D point model of the area surrounding the virtual viewpoint.