Eyes are windows to our soul
William Shakespeare

Eye Movement Data (EMD) format

EMD 1 format

Datasets A and B downloadable for competition are available in EMD1 (eye movement data) format. The format is extremely simple – it is a text file with one line for every sample. Every line is a list of tab separated elements as follows:

sid sx sy lx rx ly ry
where:

The values are 0 for point in the middle, positive for point on the right or upper side of the screen and negative for points on the left or lower side of the screen.

EMD 2 format

Datasets C and D downloadable for competition are available in EMD2 (eye movement data) format. Each recording session is presented as a file in plain ASCII format with following structure. First line of the file is a header line with description of the fields. Other line contains recorded sample information. All samples consist from 6 fields:

  1. Sequential sample number.
  2. Horizontal eye position presented in degrees of the visual angle, assuming that the center of the screen is the primary eye position, which corresponds to the beginning of coordinates. Positive values indicate right direction and negative values stand for left direction.
  3. Vertical eye position measured in degrees of the visual angle, with the same assumption that the center of the screen is the primary eye position. Positive values indicate upward direction, and negative values indicate downward direction.
  4. Data validity flag (1 for valid data and 0 for invalid data).
  5. Horizontal stimulus position measured in degrees of the visual angle, with the same rules as for eye position
  6. Vertical stimulus position measured in degrees of the visual angle, with the same rules as for eye position

For Datasets C and D. 'Training set' directory holds files that are named with subject id numbers. For example, filename id_002_1.txt means that this file was recorded from subject 2 and this is 1st recording session (session names are randomized so number does not actually mean first recorded session). And filename id_010_2.txt means that this file was recorded from subject 10 and this is 2nd recording session.

'Testing set' directory holds files that are to be associated with subject ids. These files are named task_xxx.txt, where xxx is the number of task. For example, task_001.txt mean that this is first task for identification. And task_045.txt mean that this is task number 45 for identification.

Introduction

Biometrics

Eye movements

EMD format

Competition formula

Datasets

Registration

My profile

Results (NEW)


user:

password:


(C)2012 Paweł Kasprowski