Panoramic Image Mapping Without Intermediate Distortion Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating panoramic images require significant storage resources due to the need to generate distortion-corrected and overhead images.
Innovation Solution
A method that determines pixel points in a two-dimensional planar image based on a pre-established mapping relationship with a preset stereoscopic model, allowing for the rendering of panoramic images without the need for distortion-corrected or overhead images, thereby saving storage resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If distortion-corrected images and overhead images are generated using traditional methods, then panoramic images can be obtained through texture mapping, but storage resources are consumed significantly
Solution Approach 1:
The patent extracts and eliminates the intermediate distortion-corrected images and overhead images from the traditional panoramic generation pipeline. By directly mapping pixel points from the original two-dimensional planar image to the stereoscopic model grid points, the method removes the need to generate and store these intermediate images, thereby reducing storage resource consumption while maintaining panoramic image generation capability
Solution Approach 2:
The pre-established mapping relationship serves multiple functions: it directly maps pixel points to grid points, enables rendering without intermediate images, and can be applied to different viewing angles and positions. This universal mapping approach replaces the need for separate distortion correction and overhead image generation processes
2Productivity
If traditional overhead viewpoint transformation is performed, then panoramic images can be generated, but the process requires generating and storing intermediate images
Solution Approach 1:
The patent performs preliminary action by pre-establishing the mapping relationship between pixel points and grid points before actual panoramic image generation. This pre-computed mapping table enables direct rendering without requiring complex intermediate transformations during runtime, improving generation efficiency while reducing processing complexity
Solution Approach 2:
Instead of performing complex overhead viewpoint transformations, the method uses a pre-established mapping relationship (copy of transformation data) to directly determine pixel point positions. This copying approach replaces complex real-time transformations with simple lookups, reducing both computational complexity and processing time
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
Embodiments of the present application provide a method and apparatus for generating a panoramic image. The method includes: determining, according to a pre-established mapping relationship, a pixel point in the two-dimensional planar image to which each of grid points in a preset stereoscopic model is mapped, wherein multiple grid points form one grid plane and accordingly, multiple pixel points to which the multiple grid points are mapped form one pixel region; and rendering each of the grid planes by the pixel region corresponding to this grid plane, to obtain a panoramic image. In the solution of the present application, it is not necessary to generate a distortion-corrected image and an overhead image, thereby saving storage resources.