As a newbie and low-budget neuroscience researcher, i'm looking for webcam based eye tracking softwares, which can measure ''duration of initial fixation'', ''total fixation duration'' and ''area of interest'' accurate and precisely.
I found gazerecorder, and somekind of out of date project;
http://home.agh.edu.pl/~horzyk/pracedyplom/2010SzymonDeja/2010szymondeja-pracdypl.pdf
http://blog.uxcircus.cz/eye-tracking-pomoci-standardnich-webkamer
is anyone recommend me, or used webcam tracers in their clinical research?
thank you for your answers,
Dr. Mirac
Dear Mirac,
using webcam-based eye-tracking you obviously have to make a tradeoff compared to lab-based recordings in terms of accuracy. However, the fact that most laptops nowadays have an integrated webcam, can make this type of data recording cheap and easy to achieve
Since completing my PhD in Cognitive Science last year, I am working on the Labvanced platform (https://www.labvanced.com). Recently, we integrated the webgazer.js library (https://webgazer.cs.brown.edu/ into our event system. Using this, you get an eyeposition every few milliseconds, and hence can calcualte ''total fixation duration'' and ''area of interest'' etc. in a post hoc analysis. We also extend the functionality of webgazer in a way that with the Labvanced Event System you can use fixations on any object (i.e. fixation cross) as a trigger to control the experiment (e.g. go to the next frame). Find an example webcam-based eye-tracking study here: https://www.labvanced.com/player.html?id=2909
Last but not least I can say from our experience it is important to have a good webcam, good lightning conditions, make only few head movements, as well as make calibrations between trials.
Dear Mirac,
using webcam-based eye-tracking you obviously have to make a tradeoff compared to lab-based recordings in terms of accuracy. However, the fact that most laptops nowadays have an integrated webcam, can make this type of data recording cheap and easy to achieve
Since completing my PhD in Cognitive Science last year, I am working on the Labvanced platform (https://www.labvanced.com). Recently, we integrated the webgazer.js library (https://webgazer.cs.brown.edu/ into our event system. Using this, you get an eyeposition every few milliseconds, and hence can calcualte ''total fixation duration'' and ''area of interest'' etc. in a post hoc analysis. We also extend the functionality of webgazer in a way that with the Labvanced Event System you can use fixations on any object (i.e. fixation cross) as a trigger to control the experiment (e.g. go to the next frame). Find an example webcam-based eye-tracking study here: https://www.labvanced.com/player.html?id=2909
Last but not least I can say from our experience it is important to have a good webcam, good lightning conditions, make only few head movements, as well as make calibrations between trials.
Hi Miraç Barış Usta , there is definitely a trade-off of accuracy when using a webcam compared to infrared-based eye trackers - the inevitable reduction in contrast of the eye unfortunately means that the data will likely suffer. We've written a bit about this, and have some examples of webcam software solutions here: https://imotions.com/blog/webcam-eye-tracking-vs-an-eye-tracker/ which might interest you
- Badges
- Science topic
- Similar topics
- Biological Science
- Neuroscience