• Pavis Map © 2016 IIT 5605
  • IMG 0335 1438 749x388 © 2016 IIT 5106
  • Biomedical Image Analysis Pavis © 2016 IIT 5468
  • IMG 0316 1434 749x388 © 2016 IIT 5105

PAVIS focuses on the analysis and understanding of multidimensional data like signals, images, videos and patterns in general. Its focus is on the analysis of behavior in general, grounded on computational models, targeted to the understanding of visual scenes and the investigation of brain functions (and related diseases) from neuroimaging data.

PAVIS has a wide expertise on image and signal processing, computer vision and pattern recognition, machine learning and related applications. PAVIS research involves, other than the above disciplines, also other related fields like multimedia, biometrics, multimodal data and sensor fusion, sensors networks, and embedded computer vision.

For more information please visit our site


PAVIS activities are devoted to study and to design intelligent systems for real applications, especially related, but not limited, to surveillance & security, biomedical imaging, and bioinformatics.

In particular, PAVIS addresses several specific projects such as:

  • Analysis of human behavior
  • Social Signal Processing
  • Group and crowd behavior analysis
  • Video analytics
  • Person re-identification
  • Pan-Tilt-Zoon camera sensing
  • Person detection and tracking
  • Acoustic signal processing and acoustic camera
  • Mouse behavior analysis
  • Joint structural and functional connectomics

These areas are thoroughly probed under a theoretical perspective and applied for tackling real problems, also involving other disciplinary subjects like neuroscience, psychology, biology, physics, in cooperation with research centers and universities, and both private and public partners.


The main PAVIS infrastructure consists in a 5x4 meters fully equipped anechoic chamber. This chamber allows the Department to conduct research on sound propagation and localization, as well as to perform experiments on motion capture thanks to the VICON system. This facility also supports multisensory experiments and the testing of complex signal/image processing and computer vision algorithms, concerning both 2D and 3D settings.

In particular, the lab can work with the following sensing devices:

  • Visual: 3D scanner based on Photometric Stereo, standard stereo cameras.
  • Range: Leica C10 scanner, Minolta scanner, VICON Motion Capture System, RGBD cameras
  • Audio: 12 omnidirectional lavalier microphones connected to 12 Radio Frequency transmitters; 12 Radio Frequency receivers; a 12-channel synchronous A/Ds converter




  • Prof. Shaogang Gong, Queen Mary University of London
  • Prof. Dimitri Van De Ville, EPFL
  • Prof. Xiaogang Wang, The Chinese University of Hong Kong
  • Prof. Amit Roy-Chowdhury, University California Riverside


  • NBT (L. Berdondini, F. Papaleo, V. Tucci): Joint structural and functional analysis of in-vitro neuronal networks and automatic tracking and analysis of mouse behavior
  • Nanophysics & NBT (A. Diaspro, P. Bianchini, L. Berdondini): 3D reconstruction of retina structural connectivity
  • RBCS (C. Becchio): Human behaviour analysis and prediction of intentions
  • ADVR (F. Cannella): Design of new industrial systems for visual inspection