Digital Image Editing for Folded Objects Using CNN Seam Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital image editing software struggles to create realistic images of folded objects by superimposing additional images over folds, often resulting in unrealistic outcomes due to the obscuration of image portions by folds.
Innovation Solution
The system employs machine learning algorithms, specifically convolutional neural networks, to detect seam edges in folded objects, perform image decomposition, and generate texture maps, allowing for the accurate overlay of superimposed images across folds by mapping pixels to create a realistic appearance of the superimposed image being part of the original folded object.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a second image is superimposed over the folded portion of the folded object within the original image, then the superimposed image appears on the folded object, but the resulting modified image appears unrealistic
Solution Approach 1:
The folded object image is divided into multiple unfolded image regions based on detected seam edges. Each region is processed separately to determine visibility of superimposed image portions, allowing realistic handling of folds while maintaining operational simplicity through automated region segmentation.
Solution Approach 2:
A visibility map is introduced as an intermediary data structure that stores visibility information for each pixel position across different unfolded image regions. This mediator enables the system to determine which portions of the superimposed image should be visible or obscured based on fold geometry, resolving the realism issue without complicating the user operation.
2Reliability
If conventional digital image editing is used to superimpose images on folded objects, then the process is simple, but the result is unrealistic due to improper handling of obscured portions
Solution Approach 1:
The system performs preliminary detection of seam edges and decomposition of the folded object image into unfolded image regions before superimposing the second image. This preliminary processing establishes the geometric framework needed to accurately represent folds, improving accuracy while keeping the actual superimposition operation straightforward.
Solution Approach 2:
The patent replaces manual mechanical image editing processes with automated machine learning-based seam edge detection and algorithmic visibility determination. This substitution increases accuracy in handling fold geometry while the automated nature maintains ease of operation by eliminating manual intervention.
Data Source
AI summary
Systems and techniques provide a user interface within an application to enable users to designate a folded object image of a folded object, as well as a superimposed image of a superimposed object to be added to the folded object image. Within the user interface, the user may simply place the superimposed image over the folded object image to obtain the desired modified image. If the user places the superimposed image over one or more folds of the folded object image, portions of the superimposed image will be removed to create the illusion in the modified image that the removed portions are obscured by one or more folds.