New to Social Signal Processing? Start here!
SSPNet members are not only conducting research in Social Signal Processing, but are also committed to publicizing and distributing resources aimed at lowering the entry barrier to a challenging and demanding domain like SSP.
The long term goal of our portal is to provide any interested researcher with knowledge, data and tools necessary to perform high quality research in SSP.
The portal hosts a resource repository that is at disposition to the research community and will be, as long as our portal is online, a work in progress. It will be constantly enriched with contributions from the SSPNet members, but is also open to contributions from registered portal visitors.
Use the resources menu at the top of the page to browse or search the repository and do not forget that you can contribute to them. Give visibility to your work, register and contribute to our repository with your articles, data, and software!
What are Social Intelligence and Social Signal Processing about?
Social intelligence is the facet of our cognitive abilities that guides us through the complex web of our everyday interactions, whether these require us to be a respected colleague on the workplace, a careful parent at home, a leader in our community, or simply a person others like to have around in a moment of relaxation.
At its heart, social intelligence aims at an adaptive use, accurate interpretation and appropriate display of social signals. These are discrete units of behavior, vocalization, chemicals, or morphological structure that are displayed to convey adaptive traits or states that are not directly perceivable from the outside. These can be as diverse as reproductive condition, personality traits, or attitudes towards objects or people. In everyday interactions, social signals are used to communicate interest, empathy, hostility, (dis-)agreement, flirting, dominance, superiority, inferiority, etc.
A peculiarity of social signals is that they can take the form of complex constellations of nonverbal behavioral cues (facial expressions, prosody, gestures, postures, etc.) that accompany any human-human (and human-machine) interaction. For example, laughter is a complex signal that involves changes in the facial expression, vocalization, body posture and movements of the signaller. Several decades of research in human sciences have shown that we are surprisingly efficient at understanding social signals and the variety of attitudes conveyed by them.
This leads to the three core questions addressed by Social Signal Processing:
1) Is it possible to detect automatically nonverbal behavioral cues in data captured with sensors like microphones and cameras?
2) Is it possible to automatically infer attitudes from nonverbal behavioral cues detected through sensors like microphones and cameras?
3) Is it possible to synthesize nonverbal behavioral cues conveying desired relational attitudes for embodiment of social behaviors in artificial agents, robots or other manufacts?
Most of our works revolve around these questions and involve a tight, multidisciplinary collaboration between human sciences (psychology, anthropology, sociology, etc.) on one hand, and technology (computer vision, speech analysis and synthesis, machine learning, signal processing, etc.) on the other hand. If you are interested in the same questions, this is your portal!