Eye Tracking via Iris Ellipse Fitting and Ambient Light

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional eye tracking methods for mobile devices consume excessive power due to the use of glint light sources and require expensive training data collection for machine learning techniques.

Innovation Solution

Implementing image-based eye tracking using ordinary visible light cameras, which omits the need for glint light sources, allowing for power-efficient gaze direction determination by detecting iris and pupil structures within image data, and normalizing scale and illumination to enhance detection accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If glint light sources are used for eye tracking, then gaze direction determination is enabled, but power consumption increases

Engineering Contradiction:
Improvegaze direction determinationVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent removes the glint light source component from the eye tracking system, extracting only the necessary imaging function. By using ambient light captured by the camera sensor without additional illumination, the system eliminates the power-consuming light source while maintaining gaze tracking capability through image processing of naturally lit eye structures.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system uses ambient environmental light to illuminate the eye structures being tracked, rather than requiring dedicated illumination. The camera sensor captures reflected ambient light from the eye, allowing the environment to serve the illumination function that would otherwise require active light sources, thereby reducing power consumption.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If machine learning techniques are used for eye tracking, then gaze direction accuracy is improved, but training data collection cost increases

Engineering Contradiction:
Improvegaze direction accuracyVSAvoidtraining data collection cost
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent replaces complex machine learning models requiring extensive training data with geometric and image processing methods. By using ellipse fitting to model the iris and pupil structures, and applying geometric relationships between these fitted shapes and the eye center, the system achieves accurate gaze direction determination through mathematical computation rather than data-driven learning.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system transforms the eye tracking problem from a learning-based parameter estimation task to a geometric parameter extraction task. By fitting ellipses to detected iris and pupil boundaries and computing their geometric parameters (center positions, orientations, sizes), the system derives gaze direction through deterministic mathematical relationships rather than statistical learning from training data.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If additional light sources are added to the device, then eye tracking capability is enhanced, but hardware complexity increases

Engineering Contradiction:
Improveeye tracking capabilityVSAvoidhardware configuration
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent makes the existing camera sensor perform multiple functions: it serves both as the primary imaging device for general photography and as the eye tracking sensor. By processing images from the same camera that captures regular scene photos, the system enables eye tracking without adding dedicated imaging hardware, allowing one component to fulfill multiple roles.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system extracts eye tracking functionality from a dedicated hardware subsystem and integrates it into the existing camera processing pipeline. By removing the need for separate illumination devices and specialized sensors, the patent consolidates eye tracking capabilities into the already-present camera system, reducing overall hardware complexity while maintaining tracking reliability.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP3243161B1Eye tracking
Publication Date: 2018.10.10 MICROSOFT TECHNOLOGY LICENSING LLC
  • EP3243161B1 patent drawingFigure 1~2
  • EP3243161B1 patent drawingFigure 3
  • EP3243161B1 patent drawingFigure 4

AI summary

Examples are disclosed herein that relate to eye tracking based on two-dimensional image data. One example provides, on a computing device, a method of tracking an eye. The method includes receiving image data from an image sensor, detecting a face of the user in the image data, locating the eye in a region of the face in the image data to obtain an eye image, normalizing one or more of a scale and an illumination of the eye image, fitting an ellipse to an iris of the eye in the eye image, and outputting a determination of an eye gaze direction based upon the ellipse fitted.