Eye Image Insertion Using Triangular Mesh Coordinate Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing technologies lack flexibility in inserting images into eye images, particularly in accurately determining key points and performing coordinate transformations to achieve natural integration of additional elements like eyelashes or makeup, which limits the display forms and realism of the inserted images.
Innovation Solution
A method and device that acquire eye key point data, determine auxiliary key point data, perform coordinate transformations using a moving least squares method, and insert images into specific regions of the eye image, enhancing the integration of additional elements by aligning them with the eye's geometry and opening degree, thereby improving the display form and realism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If image insertion is performed based on eye key point location only, then the insertion process is simple, but the flexibility and realism of the inserted image are insufficient
Solution Approach 1:
The patent segments the eye image into multiple triangular regions using triangulation mesh based on eye key points. This segmentation allows independent coordinate transformation and image insertion in each triangle, providing flexibility while maintaining manageable complexity through localized processing
Solution Approach 2:
The patent transforms 2D image coordinates into 3D spatial coordinates through coordinate transformation based on eye geometry. This dimensional change enables more flexible and realistic image insertion by considering the three-dimensional structure of the eye, allowing inserted images to conform to the eye's curved surface
2Manufacturing precision
If coordinate transformation is performed to achieve natural integration of inserted images, then the realism is improved, but the computational complexity increases
Solution Approach 1:
The coordinate transformation is performed separately in each triangular region rather than across the entire eye image. This segmentation reduces computational complexity by breaking down the global transformation problem into multiple smaller, independent local transformations that are easier to compute
Solution Approach 2:
The patent applies different coordinate transformation parameters to different triangular regions based on their local geometric characteristics. Each triangle receives a customized transformation that matches its specific orientation and curvature, achieving high precision integration while maintaining computational efficiency through localized processing
Data Source
AI summary
A method and device for inserting an image are provided. The method includes: acquiring a target eye image, and determining eye key point data of the target eye image; determining auxiliary key point data of the target eye image based on the eye key point data; performing coordinate transformation on the auxiliary key point data, to obtain transformed auxiliary key point data; and acquiring a to-be-inserted image, and inserting the to-be-inserted image into a region in the target eye image that is characterized by the transformed auxiliary key point data and the eye key point data.