Time and Spatial Eye-Tracking Analysis of Face Observing and Recognition
Facial images are one of the most common images in the online and offline media, which raises the question as to how people look at and remember faces. The aim of our research was to determine relation between fixation duration and saccade length in the observation process, recognition performance a...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
Faculty of Mechanical Engineering in Slavonski Brod, Faculty of Electrical Engineering in Osijek, Faculty of Civil Engineering in Osijek
2019-01-01
|
Series: | Tehnički Vjesnik |
Subjects: | |
Online Access: | https://hrcak.srce.hr/file/325806 |
Summary: | Facial images are one of the most common images in the online and offline media, which raises the question as to how people look at and remember faces. The aim of our research was to determine relation between fixation duration and saccade length in the observation process, recognition performance and furthermore to find possible relation between recognition performance and portion of observation time for internal facial features. The observation test was designed to perform time (fixations) and spatial (saccades) eye-tracking analysis when observing the facial images for three different dimensions and four different presentation times (1 second, 2 seconds, 4 seconds and 8 seconds). Recognition test showed how different presentation time and different dimensions of face image influenced recognition performance. For the recognition performance presentation time is much more important than image dimensions. Furthermore, the analysis of observing the internal facial features (eyes, nose and mouth) was also carried out. All results revealed that 4 seconds of observation time is turning point in face observation. For the 4 seconds observation test fixation duration and saccade length reached constant value, wrong recognition dropped significantly and portion of observation internal facial features reached lowest point and then increased for longer observation times. |
---|---|
ISSN: | 1330-3651 1848-6339 |