Eye Center Localization via Frontal Face Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional eye center localization methods fail to accurately locate the eye center in images with significant head rotation angles, limiting their applicability to specific frontal or near-frontal views.
Innovation Solution
An eye center localization method and system that performs image sketching, frontal face generation, eye center marking, and geometric transformation steps to calculate the eye center position information, using a processing unit and a database with a frontal face generating model and gradient method, allowing for accurate localization regardless of head rotation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional eye center localization methods are used, then the method is simple and fast, but it cannot accurately locate the eye center when the head rotation angle is too large
Solution Approach 1:
The patent introduces a frontal face generating model as an intermediary to transform non-frontal face images into simulated frontal face images. This intermediary model enables the system to handle various head positions by converting them into a standardized frontal view representation, thereby improving both localization accuracy and adaptability without requiring complex direct processing of rotated faces
2Measurement precision
If the conventional method is applied to images with large rotation angles, then the processing is fast, but the localization accuracy deteriorates
Solution Approach 1:
The patent performs preliminary action by pre-training a frontal face generating model using extensive frontal face images and their corresponding eye center positions. This pre-computed model stores the relationship between facial features and eye centers in a standardized frontal view, allowing the system to quickly transform and locate eye centers in non-frontal images without performing complex real-time calculations, thus maintaining speed while improving accuracy
Data Source
AI summary
An eye center localization method includes performing an image sketching step, a frontal face generating step, an eye center marking step and a geometric transforming step. The image sketching step is performed to drive a processing unit to sketch a face image from the image. The frontal face generating step is performed to drive the processing unit to transform the face image into a frontal face image according to a frontal face generating model. The eye center marking step is performed to drive the processing unit to mark a frontal eye center position information on the frontal face image. The geometric transforming step is performed to drive the processing unit to calculate two rotating variables between the face image and the frontal face image, and calculate the eye center position information according to the two rotating variables and the frontal eye center position information.