Panoramic Image Sky Region Masking and Virtual Object Insertion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Panoramic images shot outdoors often have a dull sky region that lacks visual interest, and existing methods do not effectively incorporate additional objects to enhance the immersive experience.
Innovation Solution
A method and apparatus that extract and mask the sky region in panoramic images, allowing for the insertion of spherical images and objects into the 3D panoramic space, enabling the creation of more engaging and diverse visual experiences by replacing the sky with celestial or dynamic images and adding virtual objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the sky region in a panoramic image is processed to create visual effects or insert objects, then the visual appeal and entertainment value are improved, but the complexity of image processing and rendering increases
Solution Approach 1:
The patent segments the panoramic image into multiple regions, specifically identifying the sky region as a separate maskable area. This allows selective processing of the sky region independently from the rest of the image, enabling visual effects and object insertion without complicating the entire image processing pipeline.
Solution Approach 2:
The patent extracts the sky region from the panoramic image by creating a mask that isolates the sky area. This extraction allows the sky region to be replaced or enhanced with virtual content while maintaining the integrity of the original panoramic structure, thus improving visual appeal without proportionally increasing processing complexity.
2Adaptability or versatility
If a mask is created to extract the sky region for replacement, then the ability to insert virtual objects is improved, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-defining the mask for the sky region and preparing the virtual object content in advance. This allows the actual object insertion process to be more efficient, as the mask and virtual content are ready to be composed into the final image without requiring complex real-time processing.
Data Source
Figure 1
Figure 2
Figure 3A~3D
AI summary
A mask region extraction unit (11) extracts a region to be masked from a panoramic image. A mask processing unit (12) generates a panoramic image in which the region from the panoramic image to be masked is masked. An alignment unit (20) aligns the direction of a celestial image with the imaging direction of the panoramic image. A mapping processing unit (14) maps the post-mask processing panoramic image and the celestial image to a 3D panoramic space as a texture. A 3D image generation unit (16) generates a 3D panoramic image which uses the imaging location as a viewpoint, and in which the 3D panoramic space is viewed in a designated line of sight.