immagine top
immagine testata
immagine finetestata
immagine iniziotestata2 immagine finetestata2
immagine shadow_1
immagine finetestata3








 


Ambient Intelligence Ambient Intelligence
Ambient Intelligence Ambient Intelligence

 News ISIP40 Open Positions
Topic: ISIP
Two PhD positions now available at ISIP40: Joint Doctoral Programme in Interactive and Cognitive Environments with the Centre for Intelligent Sensing at Queen Mary University of London

Applications are invited for 2 PhD open position

PhD
2 positions (3 years) starting in November 2018; expiration date for first level applications: May 15th, 2018 on the following research areas:
  • Self-aware cognitive dynamic systems in smart environments
  • Proactive anti-jamming in Cognitive and Software Radio

The PhD scholarships are available within the Joint Doctoral Programme in Interactive and Cognitive Environments (JD ICE) that is a PhD programme organised by the Department of Electrical, Electronic, Telecommunications Engineering and Naval Architecture at University of Genoa and the Centre for Intelligent Sensing at Queen Mary University of London (http://cis.eecs.qmul.ac.uk/) .

More details here


Posted M 12 March

 News Representation and learning of Dynamic Bayesian Interaction Models from Multisensor Data for Cognitive Self-Aware Ego-Things
Topic: ISIP
PhD Course on Representation and learning of Dynamic Bayesian Interaction Models from Multisensor Data for Cognitive Self-Aware Ego-Things

The Course will be held by Prof. Regazzoni (DITEN) on February 27-28 and on March 2-6-7 2018 at DITEN Via all'Opera Pia 11, (ex CNR building), ISIP40 Lab, 2nd floor. Lessons will be held from 10 to 13 and from 15 to 18.

Attendance is free for students from UNIGE PhD courses. However, interested attendees should send a mail by February 22nd Carlo.Regazzoni@unige.it and segr_isip@ginevra.dibe.unige.it

The course aims at providing PhD Candidates knowledge on basic state of the art and advanced theories/techniques for learning from multisensory signals and data Bayesian models for jointly predicting, processing, filtering and interpreting observed interactions. Such models will be shown to enhance functionalities of embodied smart autonomous systems like cars, radios, drones, robots, buildings by providing them a self-awareness information basis. Networks of self-aware autonomous systems interacting in smart cognitive environment will be the also targeted as examples carried on in the course. From a methodological viewpoint, this module aims at identifying and describing methodologies and techniques for defining a common probabilistic framework suitable for:

  • integrating contextual signals synchronously provided by multisensorial eso and endo sensors of autonomous systems by using Data Fusion paradigms and techniques;
  • learning from experiences behavioural and causal self-awareness models allowing an autonomous system to describe the world through a vocabulary of normal locally stationary experiences;
  • showing how each model learned from an experience can describe through probabilistic stationary rules dynamic perception, planning and actuation by means of collected external and internal observations.

Applications will be targeted of described techniques related to a couple of main case studies together with additional examples:

  • self-awareness in autonomous ground and aerial vehicles and smart infrastructures (e.g. Buildings, dynamic radio spectrum)
  • interactions in telecommunications scenarios like cognitive radio and internet of things.

Detailed Program:

  • Data Fusion methodologies and techniques for integrating multisensorial contextual data Data Fusion models.
  • JDL model and its extensions: signals, objects, situations, threats, processes and cognitive refinement.
  • Coupled Dynamic Bayesian Networks.
  • Bayesian multisensor state estimation and data association techniques:
  • Continuous and discrete state estimation techniques: from Kalman filter to Particle Filters, Hidden Markov Models.
  • PDAF and JPDAF.
  • Switching models. Markov Jump and Rao Blackwellized filters.
  • Learning methods from sparse and dense data
  • Gaussian Processes, Generative Adversarial Networks, Variational Autoencoders.
  • Unsupervised data dimensionality reduction: Self Organizing Maps, Growing Neural Gas, as semantic feature learning methods.
  • Incremental learning: Dirichlet processes
  • Techniques for non parametric self-awareness interaction-based predictive/generative/classification models .
  • Bio-inspired neural basis of consciousness: Damasio model (core-self, protoself, autobiographical memory and autobiographical self)
  • Applications and case studies:
  • Cognitive radio and Internet of Things Physical anti-jammer securitySelf-awareness in autonomous ground and aerial vehicles
  • Cognitive safety and physical security systems (smart patrolling in cooperative environments, preventive automotive vehicles, smart buidings, etc.)

Course schedule:

  • Course dates: February 27 and 28, and March 2, 6 and 7, 2018.
  • Daily lessons from 10 to 13 and from 15 to 18.
  • ISIP40 Lab - DITEN - Ex-CNR building (2nd floor) Via Opera Pia 11, 16145 Genoa.
  • For info please write an email!

Posted Friday 09 February

 News Best student paper award (silver medal)
Topic: ISIP
Paper titled "Abnormal Event Detection in Videos using Generative Adversarial Nets" by Mahdyar Ravanbakhsh, Moin Nabi, Enver Sangineto, Lucio Marcenaro, Carlo Regazzoni, Nicu Sebe won the silver medal for the best student paper award at (ICIP 2017) 2017 IEEE International Conference on Image Processing in Beijing, China

Congratulation to our PhD student Mahdyar Ravanbakhsh!
Mahdyar Ravanbakhsh receiving the award

M. Ravanbakhsh, M. Nabi, E. Sangineto, L. Marcenaro, C.S. Regazzoni, and N. Sebe, "Abnormal Event Detection in Videos using Generative Adversarial Nets", 24th IEEE International Conference on Image Processing (ICIP). 17-20 September 2017, Beijing, ChinaPDF



Abstract: In this paper we address the abnormality detection problem in crowded scenes. We propose to use Generative Adversarial Nets (GANs), which are trained using normal frames and corresponding optical-flow images in order to learn an internal representation of the scene normality. Since our GANs are trained with only normal data, they are not able to generate abnormal events. At testing time the real data are compared with both the appearance and the motion representations reconstructed by our GANs and abnormal areas are detected by computing local differences. Experimental results on challenging abnormality detection datasets show the superiority of the proposed method compared to the state of the art in both frame-level and pixel-level abnormality detection tasks.

Posted Tuesday 19 September

 News ISIP40 Open Positions
Topic: ISIP
Two PhD positions now available at ISIP40: Joint Doctoral Programme in Interactive and Cognitive Environments with the Centre for Intelligent Sensing at Queen Mary University of London

Applications are invited for 2 PhD open position

PhD
2 positions (3 years) starting in November 2017; expiration date for first level applications: May 15th, 2017 on the following research areas:
  • Probabilistic transfer learning among cognitive dynamic systems in smart environments
  • Incremental learning techniques for proactive anti-jamming in Cognitive and Software Radio

The PhD scholarships are available within the Joint Doctoral Programme in Interactive and Cognitive Environments (JD ICE) that is a PhD programme organised by the Department of Electrical, Electronic, Telecommunications Engineering and Naval Architecture at University of Genoa and the Centre for Intelligent Sensing at Queen Mary University of London (http://cis.eecs.qmul.ac.uk/) .

More details here


Posted M 27 March
 Login


:



 News
New accepted papers:

M. Ravanbakhsh, M. Nabi, E. Sangineto, L. Marcenaro, C.S. Regazzoni, and N. Sebe, "Abnormal Event Detection in Videos using Generative Adversarial Nets", 24th IEEE International Conference on Image Processing (ICIP). 17-20 September 2017, Beijing, China

D. Campo, M. Baydoun, L. Marcenaro, and C.S. Regazzoni, "Task-dependent saliency estimation from trajectories of agents in video sequences", 24th IEEE International Conference on Image Processing (ICIP). 17-20 September 2017, Beijing, China

M.O. Mughal, L. Marcenaro, and C.S. Regazzoni, "Energy Detection in Multi-user Relay Networks", Wireless Personal Communications (2017): 1-10

A. Mazzú, P. Morerio, L. Marcenaro, C.S. Regazzoni, "A Cognitive Control-inspired approach to Object Tracking", IEEE Transactions on Image Processing

V. Bastani, L. Marcenaro, C.S. Regazzoni, "Online Nonparametric Bayesian Activity Mining and Analysis From Surveillance Video", IEEE Transactions on Image Processing


 Stats
Visitors: