SocialCom12 1st International Workshop on CONTEXT BASED AFFECT RECOGNITION

 Deadline: May 11th, 2012

The first workshop on "Context Based Affect Recognition" CBAR12
 (http://contextbasedaffectrecog.blogspot.com/) will be held in conjunction with the 2012 ASE/IEEE International Conference on Social Computing SocialCom2012 (http://www.asesite.org/conferences/socialcom/2012/).

Workshop Description
The past 20 years has witnessed an increasing number
of efforts for automatic recognition of human affect using facial, vocal, body
as well as physiological signals. Several research
areas could benefit from such systems: interactive teaching systems, which
allow teachers to be aware of student stress and inattention; accident
prevention, such as driver fatigue detection; medical
tools for automatic diagnosis and monitoring such as the diagnosis of cognitive
disorder (e.g. depression, anxiety and autism) and pain assessment. However,
despite the significant amount of research on automatic affect recognition, the
current state of the art has not yet achieved the long-term objective of robust
affect recognition, particularly context based affect analysis and
interpretation. Indeed, it is well known that affect
production is accordingly displayed in a particular context, such as the
undergoing task, the other people involved, the identity and natural
expressiveness of the individual. The context tells us which expressions are
more likely to occur and thus can bias the classifier toward the most
likely/relevant classes. Without context, even humans may misunderstand the
observed facial expression. By tackling the issues of context based affect
recognition, i.e. careful study of contextual information and its relevance in
domain-specific applications, its representation, and its effect on the
performance of existing affect recognition methods, we make a step towards
real-world, real-time affect recognition.
Workshop Objectives
Context related affect analysis is still an unexplored area for
automatic affect recognition given the difficulty of modeling this variable and
of its introduction in the classification process. Unconsciously, humans
evaluate situations based on environment and social parameters when recognizing
emotions in social interactions. Contextual information helps us interpret and
respond to social interactions. 
The purpose of the workshop is to explore the benefits and drawbacks of
integrating context on affect production, interpretation and recognition. We
wish to investigate what methodologies can be applied to include contextual
information in emotion corpora, how it ought to be represented, what contextual
information are relevant (i.e. is it domain specific or not?), and how it will
improve the performance of existing frameworks for affect recognition. 
The workshop is relevant in the study of naturalistic social
interactions since contextual information cannot be discounted in doing
automatic analysis of human behavior. Embedding contextual information, such as
culture, provides a different flavor to each interaction, and makes for an
interesting scientific study. Such kinds of analysis lead us to consider
real-world parameters and complexities in affect recognition, especially in
developing human-centric systems. 
For the workshop weinvite scientists
working in related areas of affective computing, ambient computing, machine
learning, psychology and cognitive behavior to share their expertise and
achievements in the emerging field of automatic and context based affect
analysis and recognition.  
Workshop Topics
 New and unpublished papers on, but not limited to, the
following topics:
  Context source detection. 
 Context interpretation and analysis.
 Context based affect production 
 Context based facial affect recognition
 Context based vocal affect recognition
 Context based gesture affect recognition
 Context based multimodal fusion.
 Applications (Context related affect applications).

For details concerning the workshop program, paper submission guidelines, etc. please visit our workshop website at: http://contextbasedaffectrecog.blogspot.com/ 

Best regards,
Zakia Hammal

Zakia Hammal, PhD
The Robotics
 Institute, Carnegie Mellon University

Human-Machine Interaction
Facial Expression Recognition
Visual Perception


To unsubscribe from the MM-INTEREST list:
write to: mailto:[log in to unmask]
or click the following link: