2:00pm, Tuesday, 7 Aug 2007:

<center><b>
Modeling human perception of static expressions by discrete choice models
</b>

<p>
<!--<a HREF="http://itswww.epfl.ch/~sorci">-->
Matteo Sorci<!--</a>-->
<br>
EPFL

</center>


Facial expressions are probably the most visual method to convey
emotions and one of the most powerful means to relate to each other.
In this presentation we will investigate and show the use of  Discrete
Choice Models in modeling static facial expressions.
The objective of this work is focused on the behavioral  modeling of
the observer perception of human expressions.In particular we want to
investigate if differences in the observers correspond to differences
in their perception of expressions.
In order to do that we have developed a facial expression recognition
survey on the web aiming at collect data  from  a population of real
human observers, from all around the world, doing different jobs,
having different cultural backgrounds, ages and gender, belonging to
different ethnic groups, doing the survey from different places (work,
home, on travel, etc.).
Based on the collected data and on the representation of the
expressions by means of different sets of features, we have developed
and estimated three different discrete choice models.
The proposed models are then compared and validated showing
interesting and encouraging results.