Eye Movement Analysis With Semantic Segmentation for Dynamic Scenes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing eye movement analysis methods require significant time adjustments when the shape and position of research objects in scene videos change dynamically, as researchers need to manually adjust the area of interest frame by frame.
Innovation Solution
Employing a deep learning algorithm for semantic segmentation to automatically divide and identify eye movement areas of interest in scene videos, followed by superposing gaze data to obtain gaze pixel points and corresponding indices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual adjustment of area of interest is performed frame by frame, then the area of interest can cover the research object accurately, but the time required for analysis increases significantly
Solution Approach 1:
The patent replaces the manual mechanical adjustment process with an automated computer-based system that uses algorithms to automatically determine and adjust the area of interest boundaries, eliminating the need for frame-by-frame manual intervention while maintaining accurate coverage of research objects
Solution Approach 2:
The system enables automatic self-adjustment of the area of interest by using eye movement data to dynamically define and update the boundaries of the area of interest without requiring external manual intervention, allowing the system to serve itself in the adjustment process
2Extent of automation
If deep learning-based semantic segmentation is used, then automatic recognition of objects is achieved, but the device complexity increases
Solution Approach 1:
The patent integrates multiple functions into a unified system where the deep learning model serves both semantic segmentation and automatic area of interest definition, reducing overall system complexity despite the advanced algorithms used by making the system multi-functional
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The present application relates to an eye movement analysis method and system, and falls within the technical field of eye movement data analysis. The method comprises: acquiring a first scene video seen by a target user in a preset environment, and simultaneously acquiring eye movement gaze data of the target user in the environment; performing semantic segmentation on the first scene video based on a deep learning algorithm to obtain a second scene video, wherein the second scene video is divided to have eye movement area of interest; superposing the eye movement gaze data with the second scene video to obtain a gaze pixel point corresponding to the eye movement gaze data in the second scene video; and determining the gaze pixel point corresponding to each frame image in the second scene video and outputting eye movement data index of the target user gazing at the eye movement area of interest in combination with time sequence. When the shape and position of a research object in the scene video change dynamically, the present application can reduce the time required to study the eye movement behavior of the target user.