Problem
Research questions and friction points this paper is trying to address.
Binocular Gaze Estimation
Single Camera
Single Light Source
Eye Tracking
Head Movement
Binocular Gaze Estimation with Single Camera and Single Light Source
🤖 AI Summary
This work addresses the hardware constraint of conventional binocular gaze estimation methods, which typically require two light sources, by proposing a solution that operates with only a single camera and a single light source while supporting free head movement. The key innovation lies in introducing a “virtual light source” symmetric to the real one with respect to the camera, enabling the inference of a corresponding “virtual glint” position in the image through geometric relationships between the pupil and the real glint. Gaze direction is then estimated using polynomial regression under a dual-light-source assumption. Additionally, a novel normalization factor tailored for single-glint systems is designed to enhance estimation accuracy. Experimental results demonstrate that the proposed method achieves acceptable performance for lightweight applications such as mobile devices, significantly reducing hardware complexity without compromising practical usability.
📝 Abstract
According to commonly consented theories, the minimum hardware requirement for gaze tracker is one camera and two light sources to realize gaze estimation with free head movements. However, in some scenarios such as eye tracking on mobile devices, it is preferable to use less components, especially light sources. We propose a gaze estimation method with one camera and one light source. A "virtual light source" is introduced, which is geometrically placed symmetrically to the real light source with respect to the camera, and generates a "virtual glint" in the acquired image. We estimate the "virtual glint" by exploiting the relationship between the distance between two pupils and two glints in the captured image, and estimate the gaze with polynomial regression assuming two light sources are available. A new normalization factor for regression method is verified, which turns out to be practical for one-glint system. The performance is proved to be acceptable, while degradation is noticed compared to system with two actual light sources.
Innovation
Methods, ideas, or system contributions that make the work stand out.
binocular gaze estimation
single light source
virtual glint
gaze tracking
polynomial regression
🔎 Similar Papers
No similar papers found.
