Handheld Line of Sight Detection with Shake Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing line of sight detection technologies face challenges in accurately detecting a person's gaze due to camera shake and blurriness when using handheld devices like smartphones or tablets, which complicates the precision of line of sight detection.
Innovation Solution
A terminal device equipped with an imaging unit, a light source, an obtaining unit, a determining unit, and a detecting unit that assesses image blurriness caused by camera shake and only processes images deemed clear, using infrared or visible light to detect the line of sight by identifying characteristic points and correlating them with stored templates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a camera is disposed in a handheld device (smartphone or tablet), then the device portability and ease of operation are improved, but the photographed image becomes blurry due to camera shake, worsening the line of sight detection precision
Solution Approach 1:
The system performs preliminary detection of camera shake using acceleration sensors before line of sight detection. By anticipating and identifying blurry images in advance, the system can selectively process only clear images, preventing the degradation of detection precision caused by handheld operation
Solution Approach 2:
The invention introduces an intermediary detection mechanism (acceleration sensor and blur determination unit) between the camera and line of sight detection system. This intermediary layer assesses image quality and filters out blurry images, allowing the handheld device to maintain both portability and detection precision
2Productivity
If all photographed images are processed for line of sight detection, then the detection coverage is improved, but the detection accuracy deteriorates due to inclusion of blurry images caused by camera shake
Solution Approach 1:
The system applies different processing quality levels to different images based on their clarity. Clear images undergo full line of sight detection processing, while blurry images are identified and excluded or processed with reduced complexity. This local differentiation maintains high detection accuracy while preserving overall detection coverage
Solution Approach 2:
The system discards blurry images that would degrade detection accuracy, using acceleration sensor data to identify and exclude these images from processing. By selectively discarding poor-quality images rather than processing all images uniformly, the system maintains high detection accuracy while preserving productivity through efficient resource allocation
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables high-precision line of sight detection even when the device is held by a user, reducing errors from camera shake and improving accuracy by discarding blurry images and focusing on clear ones for analysis.
Implementation Method 1
causes a light source 15 to emit an infrared ray to irradiate the user P1... generates an image by photographing the user P1 irradiated by the infrared ray
Implementation Method 2
The imaging unit 10 generates an image by photographing the user P1
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
A terminal device includes a computer processor, the device includes: an obtaining unit configured to obtain information indicating a state of an imaging unit when photographing a subject; a determining unit configured to determine whether an image photographed by the imaging unit includes blurriness due to shaking of the imaging unit based on the obtained information; and a detecting unit configured to use an image determined as not including blurriness to detect a line of sight of the photographed subject.