Panoramic Image Stitching Using Depth Map Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current panoramic image generation techniques often result in mis-registrations and inefficient seams when stitching images, leading to low-quality panoramic images due to the lack of effective registration and blending methods.
Innovation Solution
The use of depth map data for registration, labeling, and blending of images, where images are divided into depth planes to calculate registration matrices and identify seams, allowing for the creation of high-quality panoramic images by connecting and blending image portions across these planes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If conventional registration and blending tools are used to stitch images, then the process can be completed, but mis-registrations and inefficient seams occur resulting in low quality panoramic images
Solution Approach 1:
The patent divides images into multiple depth planes, where each plane contains pixels belonging to a specific depth range. This segmentation allows independent processing of different depth layers, enabling accurate registration and blending for each plane while maintaining overall image quality. The depth planes are processed separately through registration matrix calculation and seam identification, resolving the contradiction between processing complexity and output quality.
Solution Approach 2:
The patent introduces a depth dimension to traditional 2D image stitching by creating multiple depth planes. This adds a third dimension (depth) to the registration and blending process, allowing algorithms to account for spatial relationships between objects at different distances. The depth map data provides additional information that improves registration accuracy and seam efficiency without compromising panoramic image quality.
2Ease of manufacture
If images are stitched together using traditional methods, then panoramic images can be generated, but seams are inefficient and visible resulting in undesirable quality
Solution Approach 1:
The patent applies different processing techniques to different regions of the image based on their depth characteristics. Each depth plane is processed independently with its own registration matrix and seam identification, allowing optimal blending parameters to be applied locally. This ensures that seams are efficiently blended in regions where they occur while maintaining image quality in other regions, resolving the contradiction between ease of generation and blending precision.
3Measurement precision
If depth map data is used to divide images into multiple planes, then registration accuracy improves, but processing complexity increases
Solution Approach 1:
The patent introduces depth map data as an intermediary that mediates between the input images and the final panoramic output. The depth map serves as an intermediate representation that organizes pixels into depth planes, providing structured information that simplifies the registration and blending process. Rather than directly processing complex 2D image relationships, the system uses depth map data as a mediator to guide the stitching process, improving precision while managing complexity through this intermediate layer.
Data Source
AI summary
An apparatus for generating a panoramic image(s) may include a processor and memory storing executable computer code causing the apparatus to at least perform operations including generating a panoramic image(s) based on performing registration on at least two images associated with depth map data and determining an overlap region(s) between the two images within first and second items of depth map data. The computer program code may further cause the apparatus to label the overlap region(s) within the first or second item of depth map data based on generating a virtual seam(s) connecting portions of the two images to generate a composite image. The portions correspond to image data within first and second items of depth map data. The computer program code may further cause the apparatus to blend the seam of the composite image to obtain the panoramic image. Corresponding methods and computer program products are also provided.