SIMBAD -- Beyond Features: Similarity-Based Pattern Analysis and Recognition (project no. 213250), funded by European Commission, FP7-ICT-2007-C FET Open (7th framework).
SIMBAD aims at undertaking a thorough study of several aspects of purely similarity-based pattern analysis and recognition methods, from the theoretical, computational, and applicative perspective. SIMBAD aims at covering a wide range od problems and perspectives. SIMBAD will consider both supervised and unsupervised learning paradigms, generative and discriminative models, and SIMBAD's interests will range from purely theoretical problems to real-world practical applications.

SAMURAI -- Suspicious and Abnormal behaviour Monitoring Using a netwoRk of cAmeras & sensors for sItuation awareness enhancement
The aim of SAMURAI is to develop and integrate an innovative intelligent surveillance system for robust monitoring of both inside and surrounding areas of a critical public infrastructure site (e.g. an airport concourse, underground platforms). SAMURAI has three significant novelties that make it distinctive from other recent and ongoing relevant activities both in the EU and elsewhere:

• SAMURAI is to employ networked heterogeneous sensors rather than CCTV cameras alone so that multiple complementary sources of information can be fused to create a visualisation of a more complete .big picture. of a crowded public space.

• Existing systems focus on analysing recorded video using pre-defined hard rules, suffering from unacceptable false alarms. SAMURAI is to develop a real-time adaptive behaviour profiling and abnormality detection system for alarm event alert and prediction with much reduced false alarm.

• In addition to fix-positioned CCTV cameras, the SAMURAI system will also take command input from control room operators and mobile sensory input for patrolling security staff for a hybrid context-aware based abnormal behaviour recognition. This is in contrary to current video behaviour recognition system that relies purely on information extracted from the video data, often too ambiguous to be effective.

3-SHIRT -- Three-dimensional SHape Indexing and Retrieval Techniques , partially funded by the Italian Ministry of Education, University and Research
The project aims at developing innovative solutions in all the phases of content-based 3D shape retrieval, namely: shape analysis and segmentation, design of shape descriptors, shape indexing and matching, and evaluation. Content-based shape retrieval consist of three main steps: descriptors extraction, indexing and matching. 3-SHIRT will address all these aspects, including also shape analysis (pre-processing) and overall evaluation. The issue of shape analysis consist of several techniques for the semantic representation of three-dimensional objects (models), like 1D skeletonization, features extraction, and segmentation. One of major challenges is to elaborate a suitable canonical characterization (descriptor) of the entities to be indexed. Since the descriptor serves as a key for the search process, it decisively influences the performance of the search engine in terms of computational efficiency and relevance of the results. 3-SHIRT aims at devising new descriptors characterized by being i) invariant to more complex transformations ii) robust to occlusions iii) more closely related to human perception (and to human expectations) and iv) by including texture information as well as geometry. As for the matching phase, 3-SHIRT will focus on the design of matching methods with particular emphasis on the statistical learning approach for modelling the object's class variability.

Similarity-based Methods for Computer Vision and Pattern Recognition: Theory, Algorithms, Applications , partially funded by the Italian Ministry of Education, University and Research
In this project we aim at undertaking a thorough study of several aspects of similarity-based pattern recognition methods, thereby advancing the state of the art in the field. Specifically, we shall focus primarily on problems and algorithms involving non-vectorial data and representations such as relational structures (e.g. graphs) and graphical models (e.g., hidden Markov models), whereby similarity-based methods best exhibit their power and flexibility over traditional feature-based approaches. The research project aims at covering a wide range of problems and perspectives. We shall consider both supervised and unsupervised learning paradigms, generative and discriminative models, and our interest will range from purely theoretical problems to real-world practical applications. To this end, the research network has been carefully designed so as to include high-profile competences in all the areas considered. In addition, we strove to set up a highly cohesive network where, not only each research unit is well characterized in terms of competences, problems addressed, methodologies used and objectives, but is also tightly coupled with the rest of the network.

Closed -- Closing the Loop Of Sound Evaluation and Design (FP6-NEST-PATH "measuring the impossible" project no 29085).
Despite being a promising and lively playground, sound design is not a solid discipline yet. We believe that the reason is to be found in the lack of design-oriented measurement and evaluation tools. The CLOSED project aims at providing a functional-aesthetic sound measurement tool that can be profitably used by designers. At one end, this tool will be linked with physical attributes of sound-enhanced everyday objects; at the other end it will relate to user emotional response. The measurement tool will be made of a set of easy-to-interpret indicators, which will be related to use in natural context, and it will be integrated in the product design process to facilitate the control of sonic aspects of objects, functionalities, and services encountered in everyday settings. The aim of the CLOSED project is to provide such concepts and tools, toward closing the loop of sound evaluation and design.
Duration: July 1, 2006-June 30, 2009.
FP6-NEST-PATH "measuring the impossible" project no 29085.

Sound Synthesis By Physical Models Of The Piano
The two competitive physics-based synthesis models of the piano, the one based on digital waveguides, and the other based on modal decomposition, will be accurately compared in terms of efficiency, timbral quality, and versatility. Given the software architecture of choice, specific signal processing blocks will be developed to realize the model constituting blocks, each block being dedicated to a specific mechanism (hammer-string interaction, transversal vibrations, longitudinal modes, soundboard, mode beatings, etc.), following a functional decomposition of the piano. The realization will take into account the constraints deriving from a hypothetical final implementation on a Digital Signal Processor, thus considering the requirements of an industrial exploitation of the model. The final part of the project will be dedicated to the parameterization and tuning of a playable version of the model, that is a plugin for a sound processing software platform.
Duration: 2007-2009 (24 months)

Genetic Bases of β-Cell Role in Glucose Homeostasis of Patients With Type 2 Diabetes: A Computational Biomedicine Study.
Granted by the European Foundation for the Study of Diabetes (EFSD) and Novartis as part of the European Program for Clinical Research in the Field of Pancreatic Islet Dysfunction (applicant Prof. Riccardo Bonadonna, of the Endocrinology Division at the Department of Biomedical and Surgical Sciences of the University of Verona).
Duration: Summer 2008-Summer 2010.
Grnated by EFSD and Novartis.

Industrial funding.
This work proposes an efficient acoustic field visualisation technique for the monitoring and the surveillance of a large sea area. The method does not require active sensing instrumentation as it is based on a simple passive acoustic approach. The visualisation allows an operator to obtain a better perception of the acoustic field and a prompt localisation of extraneous entities in the monitoring area.

LIMA 3D -- Low cost 3D imaging and modelling automatic system, partially funded by the Italian Ministry of Education, University and Research.
LIMA 3D is a COFIN project. The project aims at the development of a low-cost system for the three-dimensional (3D) modeling of scenes and objects starting from the acquisition of real subjects.
From analysis and processing of raw data (measured by the sensors), 3D models will be extracted and utilised for different applications. The system will be focused on all the modeling stages, from the sensor design and development, and related image generation (optical and range), to the graphic rendering and synthetic reproduction of the observed objects and scenes. The system, and the specific developed techniques, will be characterised by the versatility of dealing with different types of data and will be devoted to the reconstruction of both single objects and whole scenes.

ARROV -- Augmented Reality for Remotely Operated Vehicles based on 3D acoustical and optical sensors for underwater inspection and survey (project no. GRD1-2000-25409), funded by European Commission, Sustainable and Competitive Growth key action (5th framework).
Improving the applicability and performance of multifunctional Remotely Operated Vehicles (ROV) for the inspection, monitoring and survey of underwater environments and manmade installations. This goal will be achieved by developing a novel integrated sensorial system, based on both optical and acoustical data, to support the ROV pilot with an augmented representation of the scene (augmented reality).
This project is a follow-up of a BRITE-EURAM 3 Basic Research project named VENICE (Virtual ENvironment interface by sensory integration for Inspection and manipulation Control in multifunctional underwater vehicles).

Industrial funding.
The objective of this project is to develop a technology (algorithms and prototype) thet will enable blind people to see the world with their ears. By equipping the blind people with a portable computer, earphones and a stereo camera pair, it is possible to give him a thee-dimensional (3D) reconstruction of the scene that is encoded by appropriate sound syntesis techniques.

SOb -- The Sounding Object (project no. IST-2000-25287), funded by the European Commission, part of the Disappearing Computer proactive initiative of the IST Future and Emerging Technologies.
The SOb project aims at developing sound models that are responsive to physical interactions and are easily matched to physical objects. The sound models, being specified by pysical descriptions and actions, will be ready to be integrated into artefacts that interact with each other and that accessed by direct manipulation. Control models are developed in order to reproduce parametric variations that are natural according to the dynamics of human gestures and expressive intentions. Sound and control models are developed after the phenomenological and psychophysical characterisation of a restricted class of sound events. The results of research are demonstrated by means of a dynamic sound library and an application that will allow users to interact with objects using only gestures and auditory display.

S2S² -- Sound to Sense, Sense to Sound (FET-Open Coordination Action, number of contract: IST-2004-03773).
S2S² (acronym for ''Sound to Sense, Sense to Sound'', pronounce s two - s square) has the following overall objective: to bring together the state-of-the-art research in the sound domain and in the proper combination of human sciences, technological research and neuropsychological sciences that does relate to sound and sense. Reaching this objective can foster a new generation of research topics such as higher-level sound analysis, the so-called ``engaging'' synthesis, an integrated sound-music research field, etc. The coordination action S2S² has been conceived to prepare the scientific grounds on which to build the next generation of scientific research on sound and its perceptual/cognitive reflexes. So far, a number of fast-moving sciences ranging from signal processing to experimental psychology, from acoustics to cognitive musicology, have tapped the S2S² arena here or there. What we are still missing is an integrated multidisciplinary and multidirectional approach. Only by coordinating the actions of the most active contributors in different subfields of the S2S² arena we can hope to elicit fresh ideas and new paradigms.
Duration: June, 1 2004-May 31, 2007
FET-Open Coordination Action (number of contract: IST-2004-03773;).

SCoCoDe -- Sound/Control Co-Design: sound synthesis and devices for gesture based control
The main goal of the project is to establish an effective design process, in which the development of algorithms for sound synthesis and processing is driven by control requirements and, on the other hand, the development of gesture-based control models is based on the inherent physical interpretation of the sound model parameters. To this end, it seems appropriate to adopt the physically-based approach, in which the sound synthesis algorithms are obtained by simulating the objects and the physical interactions that are responsible for the acoustic events: using this approach, the control parameters can be mapped in a natural way into physical quantities, thus helping the design of the control layer. Control is achieved through the development of algorithms for recognition and processing of the physical/geometrical information and of the expressive content communicated by a user through his gestures: effective strategies to use such information in order to control the synthesis/processing layer have to be found and specified. The effectiveness of the design process proposed in this project will be demonstrated through the development of two prototypal applications, both focused on the interaction of the user's gestures and movement with the auditory feedback.
Duration: 2002-2004 (24 months)
MIUR: Programmi Di Ricerca Scientifica Di Rilevante Interesse Nazionale.

Synthesis of interactive sounds from relevant ecological phenomena
Sound production with the computer relies on well-established methods for the synthesis of discrete-time signals which are informative from an auditory point of view, and on some subsequent computer-based discrete-time processing. The impulse toward the research of new methods, once mainly dictated by the suggestions coming from the field of music production, today is influenced with increasing strength by models coming from cognitive acoustics, auditory perception and human-computer interaction.
This project aims at achieving three main goals:
1. Design and further development of an general-purpose impact/friction model devoted to the synthesis of impulse and sustained sounds, provided also by a number of control methods for the reproduction of different interaction contexts.
2. Design and further development of a spatialization model devoted to provide sounds with the information about angular position and distance between source and listener.
3. Definition of an environment for real-time model execution.
Duration: 2002-2004 (24 months)
MIUR: Programmi Di Ricerca Scientifica Di Rilevante Interesse Nazionale.