Eye movement detection project

Final project by

Nir Arcushin and Ellen Plosker

Introduction

What captures our attention as humans when we look at different pictures?Is there any part of the picture is most probable to capture our mind?What happens when we see a face on it?

The answer to those and many other questions could be found in experiments using the EyeLink II video-based eye tracking system developed bySR Research to measure human eye movements. As shown in the figure below, it uses a headsetwith infrared cameras mounted on it to obtain information on a subject’s current gaze position.

Head Mounted System:

The system consist of two PCs:
The display PC, running Windows XP, is used to display pictures and collect responses from the subject.
The host PC, running ROMDOS, records the data streams from the eyetracker.
We also had a license for the EyeLink Data Viewer developed by SR Research for viewing, filtering, and outputting EyeLink data events.

Our first goal was to provide a prototype of an eye movement trackingsystem optimized for detecting eye's gazes at serials of pictures.
The second goal was to program and run an experiment whose results could provide us an answer to the third question we asked : "What happens when we as humans see a human face on a picture?"

Approach and Method

The project divided into 2 parts:

Part 1 :
Integration and configuration of the system using supplied software from SR Research Data Viewer.


Part 2:
Understanding provided templates written in C and programming an experiment using those templates.In experiment the subject exposed to serias of pictures which include faces of different people. His eye movements were recorded and introduced to the researcher as the original picture with stamps of gazes of the subject.

Results

The following list of pictures were introduced to the subject :


Another interesting picture:

The pictures were chosen, so that the faces will appear in different sizes and different positions on the image and displayed to the subject each picture for 2 seconds. The output of Data Viewer after the experiment is stamps of gazes on the original picture.

And the output of the Data Viewer :


And the "picture with many faces" :

Conclusions

As you can see in the output images the points of gazes are very close to the faces on the pictures, even that there are other interesting and colorful parts of image. There is some small inaccuracy in gaze points because of sensitivity of the headset.So disregarding this small inaccurancy we can see that indeed faces capture our mind more then other part of image.

Additional Information

References

SR Research User Manuals:

ELII Head Mounted Installation Guide V3.02.pdf
EyeLink Data Viewer User Manual.chm
EyeLink Programmers Guide (SDL).pdf
EyeLink API Specification.pdf
EyeLink II Head Mounted User Manual 2.12.pdf