[Xmca-l] Labvanced.com New Eye-Tracking Cutting Edge Technology

[Leah Lee]

*

Hello everyone,


Labvanced <https://urldefense.com/v3/__https://www.labvanced.com/__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkB1j8HBrw$ >has just released a cutting 
edgedeep learning based eye tracking algorithm,which uses the webcam to 
track the gaze of participants. We are highly confident that not only is 
this much better than webgazer.js or other algorithms doing webcam based 
eye tracking, but we actually believe that the accuracy is good enough 
for at least 80-90% of eye tracking research, which is normally done in 
the lab. We made several demos, so please try and give us feedback about 
this approach. You can use these studies also as templates for your own 
eye tracking research onLabvanced 
<https://urldefense.com/v3/__https://www.labvanced.com/__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkB1j8HBrw$ >(https://urldefense.com/v3/__https://www.labvanced.com__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkBmLfgChg$  
<https://urldefense.com/v3/__https://www.labvanced.com/__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkB1j8HBrw$ >).  Please note that this algorithm runs 
completely client side and no video data / face data will be stored on 
our system, so your privacy and your participants’ privacy is guaranteed!


 1.

    https://urldefense.com/v3/__https://www.labvanced.com/page/library/11099This__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkAdhIE_vg$  is a performance
    test where the predicted gaze position will be displayed onto the
    screen, so you can test how good the prediction really is.

 2.

    https://urldefense.com/v3/__https://www.labvanced.com/page/library/11327This__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkAXm8JUmA$  is a little game
    where you have to fixate objects to "destroy" them before they reach
    the bottom of the screen.


A few things to note:

  *

    Calibration will currently take 7-8 minutes. We will cut this down
    over time to about 4-5 minutes.

  *

    It requires a decent computer/laptop, we will soon provide an
    initial “system check” to exclude people with too old equipment, or
    other reasons why it might not work.

  *

    Further improvements in accuracy will most likely be coming in the
    next months by leveraging larger training sets and improving drift
    correction.


  If you have any questions about our new Eye-Tracking technology, or if 
you are interested to use our platform, please email me, we will also be 
happy to provide a free demo via Google Meet at your convenience.


Best regards,

Leah Lee

*
--
Labvanced <https://urldefense.com/v3/__http://www.labvanced.com__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkAqBav46g$ > - The intuitive cloud solution for 
online experiments

Grunigerstr. 19
33102 Paderborn
Germany

phone: (+49) 5251-5449385
website: https://urldefense.com/v3/__http://www.labvanced.com__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkAqBav46g$  <https://urldefense.com/v3/__http://www.labvanced.com/__;!!Mih3wA!UpzqleXFRi_xUtWuWg3gQkBBGbszk7m3q3kCjHRdybKo3JVvZkGjx0_QVCxvgkAod2MzDg$ >
email: leah@labvanced.com <mailto:leah@labvanced.com>

Scicovery GmbH Amtsgericht Paderborn HRB 13448
Geschäftsführer: Caspar Goeke, Holger Finger
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucsd.edu/pipermail/xmca-l/attachments/20200807/36300385/attachment.html