The Summer School on Signal Processing for Interactive and Cognitive Environments (SP-ICE) is the second edition of the Summer Schools series related with the Erasmus Mundus Joint Doctorate on Interactive and Cognitive Environments (EMJD-ICE http://www.icephd.org) and can be considered as an ideal prosecution of the first edition that was organized by the Alpen-Adria Universität in Klagenfurt (AT) and technically co-sponsored by IEEE via the Seasonal Schools in Signal Processing (S3P) initiative.

Main topics of the Summer School activities are machine learning, cryptography, video-processing including tracking and detection, ambient assisted living, facial biometrics, emotion recognition, serious gaming, distributed consensus, bio-inspired systems, human interactions recognition.

1 Research grant position is available at ISIP40 group at the University of Genova on the following topic:

• Novel algorithms and methodologies for fusing heterogeneous data from object tracking for understanding interactions between moving subjects within a guarded environment (Research grant)

  The main objective of the research activity will be to study and develop methodologies and algorithms inspired by human reasoning being able to analyze contextual data from multiple heterogeneous sensors. A tool will be developed able to acquire multiple trajectories of heterogeneous objects within an external environment and produce a) a structured information about monitored situations indicating anomalous events and b) information about possible decisions that should be able to revert the system to a normal working condition.

Monday, March 26, 09:30 - 12:30
T-7: Distributed Data Fusion for Interactive Cognitive Environments
Carlo Regazzoni, Lucio Marcenaro
Last version of the tutorial presentation is available from here.

• T-7: Distributed Data Fusion for Interactive Cognitive Environments
  Carlo Regazzoni, Lucio Marcenaro

  The tutorial aims at providing an overview of data fusion techniques necessary for representing, modeling and automatically interpreting complex interaction situations occurring in interactive cognitive environments starting from observations provided by a distributed network of embedded systems interacting with the observed environment. A Bayesian oriented viewpoint is assumed in this tutorial.

  The main objective is to describe a common framework based on distributed data fusion, for multi-level joint tracking and classification of objects and interactions. The distributed data fusion theory is applied to the problem of recognizing multiple interacting objects from multi-camera video sequences. In particular, the attention will be focused on those actions involving more than one subject, or a subject and elements in the environment, i.e. interactions. This approach is quite novel in the domain of human activity recognition since most methods rely just on the analysis of body motion of single subjects but they don't take into account the context where these situations take place. To this goal, video processing algorithms aiming at improving tracking performances will also be described, both for detection of subjects and body parts. Collaborative trackers, being designed to exchange information between each other in order to produce more reliable models, are a promising approach to tackle typical issues due to occlusions happening during interaction among multiple subjects.

  The tutorial will demonstrate how the same distributed data fusion theories can be successfully applied at a higher level for classifying interactions between subjects on a larger scale. Object classification is a fundamental step in automatic video tracking which allows improved tracking and a more accurate description of events. The hypothesis for object classification is based on feature selection method which gives a good subset of features while the machine learns the classification task and use these selected features for object classification.

  The tutorial is divided into the following parts: Introduction on Interactive and Cognitive Environments (C.Regazzoni) Data fusion architectures and models (L.Marcenaro) Probabilistic Graphical Models and Dynamic Bayesian Networks for interactions modeling (C.Regazzoni) Examples and case studies (L.Marcenaro)

  Target audience:
  This tutorial is specifically intended for researchers studying signal processing in interactive and cognitive environments. Data fusion plays an important role for developing this kind of systems. Typical audience of this tutorial comprehends experts studying methods that involve distributed data sources from small to large sensors. The audience should have a working knowledge of mathematics and probability theory. Some familiarity with data fusion and tracking is desirable but not required.

ISIP40 research group is one of the organizers of the second edition of the GTTI Thematic meeting on Multimedia Signal Processing that will be held in Courmayeur (AO) on March 19-21, 2012.

This GTTI research area was established in 2011 with the first thematic meeting organized by the Trento University Research Unit in Folgarida (TN). In 2012, the second thematic meeting is organized by the University of Genova.

The main program of the meeting includes a Demo and Exhibit Session. Demonstrations are intended as a forum for showcasing the state of the art in Multimedia Signal Processing. The focus will be on innovative prototype implementations, systems and technologies in this research field. Demos can be considered as real, practical, and interactive proof of the presenters' research ideas and scientific or engineering contributions, with the goal of providing multimedia researchers and practitioners with the opportunity to discuss working multimedia systems, applications, prototypes, or proof-of-concepts. Such a setting allows meeting attendants to view and interact first hand with live evidence of innovative solutions and ideas in the field of multimedia, and to see leading edge research at work. Demonstrators will be provided with a table, power outlet, and (shared) wireless Internet. Demo presenters are expected to bring with them everything else needed for their demo, such as hardware, laptops, sensors, PCs, etc.

Submission Information

Submissions are encouraged in all areas related to multimedia signal processing. Please submit short demo descriptions (max 1 page in English or Italian) to gtti.mmsp2012@dibe.unige.it describing the scientific or engineering concept behind the work, the novelty of the work, how is the work different from existing systems/techniques, and what will be actually shown during the demo..

Best Demonstration Award

All research demo submissions with a student lead will be automatically entered into a Student Research Demo competition. During the meeting, demos will be judged by an evaluation committee and one will be selected for the Best Demonstration Award. Demos will primarily be judged on originality, execution, and illustration of the core research ideas. The 500€ prize is sponsored by GTTI.

For any question about the meeting please write to gtti.mmsp2012@dibe.unige.it or follow this link.