Eye Tracking via PCA Image Reconstruction and Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 3D head-up display (HUD) systems face challenges in accurately tracking eye positions, especially in environments with inadequate lighting or when noise components like sunglasses or light reflections interfere, making it difficult to maintain effective eye tracking.
Innovation Solution
An eye tracking method that generates a reconstructed image through noise reduction using principal component analysis, determines a difference value between the input and reconstructed images, and selects the most suitable image (input, reconstructed, or replacement) based on thresholds to perform eye tracking, ensuring accurate eye position detection even in challenging conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional eye tracking methods are used in challenging environments, then the system complexity remains low, but eye tracking accuracy deteriorates due to noise and occlusions
Solution Approach 1:
The system performs preliminary noise reduction and image reconstruction before eye tracking to improve accuracy in challenging environments. By preprocessing the input image through PCA-based reconstruction and comparing multiple image versions (original, reconstructed, and replacement), the system prepares optimized images in advance for more reliable eye position detection.
Solution Approach 2:
The system introduces an intermediary image reconstruction process that acts as a mediator between the raw input image and the eye tracking algorithm. The reconstructed image serves as an intermediate representation that reduces noise components while preserving eye-related features, and the system selectively uses either the original or reconstructed image based on which provides better tracking results.
2Reliability
If noise reduction through PCA is applied to improve image quality, then eye tracking reliability improves, but processing time increases
Solution Approach 1:
The system applies partial noise reduction by using PCA to reconstruct images with a limited number of principal components rather than complete decomposition. This partial action approach reduces processing time while still capturing the most significant variance in the image data, achieving a balance between reliability improvement and time efficiency.
Solution Approach 2:
The system dynamically adjusts the image processing approach based on real-time conditions. It compares the original input image with the PCA-reconstructed image and selectively uses the version that provides better eye tracking results. This dynamic adaptation allows the system to increase processing effort only when necessary, minimizing time loss while maintaining reliability.
3Measurement precision
If multiple image versions are compared and selected based on thresholds, then eye tracking accuracy improves in various conditions, but computational complexity increases
Solution Approach 1:
The system segments the image processing into distinct versions: the original input image, the PCA-reconstructed image, and replacement images from a database. By dividing the processing into these segments and comparing them using threshold-based selection, the system achieves comprehensive coverage of different lighting and noise conditions without requiring a single overly complex processing pipeline.
Solution Approach 2:
The system changes the parameter of image representation by generating multiple versions with different processing levels (original, PCA-reconstructed, and database replacements). It then selects the appropriate version based on threshold comparisons of difference values, effectively adapting to varying environmental conditions through parameter changes rather than complex real-time analysis.
Data Source
AI summary
An eye reconstruction-based eye tracking method and apparatus are provided. The eye tracking method includes generating a reconstructed image by performing eye reconstruction with respect to an input image, determining a difference value between the input image and the reconstructed image, selecting one of the input image, the reconstructed image, and a replacement image as a target image based on the determined difference value, and performing eye tracking based on the target image.