Program

This is the preliminary GTTI MMSP 2016 schedule, more details and updates will be added as soon as they will be defined. Click on a specific event to explore the details and to add it to your personal agenda.


Technical Program 

Monday, January 25th

Keynote Speech: Learning Sparsifying Transforms for Signal, Image, and Video Processing

Yoram Bresler (Coordinated Science Laboratory and the Department of ECE, University of Illinois at Urbana-Champaign)

Time: 14:15 - 15:15
Abstract: The sparsity of signals and images in a certain transform domain or dictionary has been exploited in many applications in signal and image processing, including compression, denoising, and notably in compressed sensing, which enables accurate reconstruction from undersampled data. These various applications used sparsifying transforms such as DCT, wavelets, curvelets, and finite differences, all of which had a fixed, analytical data-independent form. Likewise, the acquisition in compressed sensing used mostly random sparse sampling schemes, chosen in a universal way, independent of the data.

Recently, sparse representations that are directly adapted to the data have become popular, especially in applications such as image and video denoising and inpainting. While synthesis dictionary learning has enjoyed great popularity and analysis dictionary learning too has been explored, these methods involve a repeated step of sparse coding, which is NP hard, and heuristics for its approximations are computationally expensive. In this talk we describe our work on an alternative approach: sparsifying transform learning, in which a sparsifying transform is learned from data. The method provides efficient computational algorithms with exact closed-form solutions for the alternating optimization steps, and with theoretical convergence guarantees. The method scales better than dictionary learning with problem size and dimension, a in practice provides orders of magnitude speed improvements and better image quality in image processing applications. Variations on the method include the learning of a union of transforms, and online versions.

We describe applications to image representation, image and video denoising, and inverse problems in imaging, demonstrating improvements in performance and computation over state of the art methods.

Session 1: Compression of Multimedia Signals

Session Chair: Enrico Magli (Politecnico di Torino)
Time: 15:15 - 16:30
Program:

  • Local feature coding for distributed visual analysis, Luca Baroffio, Luca Bondi, Matteo Cesana, Alessandro Redondi, Marco Tagliasacchi, Stefano Tubaro (Politecnico di Milano)
  • Compression of photo collections using geometrical information, Simone Milani, Pietro Zanuttigh (Università degli Studi di Padova)
  • Graph-based transforms for image and video compression, Giulia Fracastoro, Enrico Magli (Politecnico di Torino)
  • Business and technology: video codecs and IPR licensing, Paola Sunna (RAI CRIT)

Session 2: Image Analysis and Retrieval

Session Chair: Riccardo Leonardi (Università degli Studi di Brescia)
Time: 16:50- 18:00
Program:

  • License Plate Recognition with Fully Convolutional Neural Network, Mauro Annarumma (Telecom Italia)
  • Saliency-based event recognition in visual media, Francesco De Natale (Università degli Studi di Trento)
  • Image and geometry processing in a 3D-modeling perspective, Alberto Signoroni (Università degli Studi di Brescia)
  • Depth Field Estimation in Plenoptic Cameras, Alessandro Neri, Marco Carli, Federica Battisti (Università degli Studi Roma Tre)


Tuesday, January 26th

Keynote Speech: Graph Signal Processing: Uncertainty Principle and Sampling Theory

Sergio Barbarossa (University of Rome "La Sapienza")

Time: 09:30 - 10:30
Abstract: In many applications of current interest, the observations can be represented as a signal defined over a graph. The analysis of such signals requires the extension of standard signal processing tools. After reviewing the more recent contributions in this field, this talk describes alternative ways to define a Graph Fourier Transform (GFT). Then, building on the GFT, it illustrates an uncertainty principle for signals on graph and shows how to build a dictionary of maximally concentrated signals on the graph and dual domain. A direct relation between uncertainty principle and sampling theory is then established. This theory illustrates alternative mechanisms to sample a graph signal and to recover the original signal from its samples. It is shown how, when sampling a graph signal, it is not only important to known how many samples to take but also where to take the samples, as their location plays a key role in the performance of the recovery algorithms. Finally, the conditions for perfect recovery of a useful signal corrupted by sparse noise are illustrated, showing that this problem is also intrinsically related to graph signal localization properties.

Demos & Posters Session

Session Chair: Pietro Zanuttigh (Università degli Studi di Padova)
Time: 10:50 - 13:00
Demos:

  • 3D Puzzles: a Gaming Approach for Cognitive Rehabilitation, Mattia Dal Pont, Nicola Conci, Francesco De Natale (Università degli Studi di Trento)
  • A Hierarchical Bayesian Model for Emotion Estimation in Crowded Environments, Oscar J. Urizar, Mirza Sulman Baig, Lucio Marcenaro and Carlo S. Regazzoni (Dipartimento di Ingegneria Navale, Elettrica, Elettronica e delle Telecomunicazioni (DITEN), Università degli Studi di Genova)
  • Embedded Activity Recognition: a Prototype, Danilo Pau, Matteo Cesana (STMicroelectronics) (video demo with poster)
  • Head-Mounted Gesture Controlled Interface for Human-Computer Interaction, Alvise Memo, Ludovico Minto and Pietro Zanuttigh (Università di Padova)
  • Identification and Analysis of Static Characteristics of Environments from the Motion of Agents, Damian Campo, Lucio Marcenaro and Carlo S. Regazzoni (Dipartimento di Ingegneria Navale, Elettrica, Elettronica e delle Telecomunicazioni (DITEN), Università degli Studi di Genova)
  • Interactive Augmented Reality Using a Projector-Depth-Camera System, Simone Milani (Università degli Studi di Padova)
  • License Plate Recognition with Fully Convolutional Neural Network, Mauro Annarumma, Gianluca Francini (Telecom Italia), Tomas Bjorklund and Enrico Magli (Politecnico di Torino)
  • Multiple Instances Visual Search, Marco Paracchini, Marco Marcon (Politecnico di Milano) Emanuele Plebani and Danilo Pau (STMicroelectronics) (video demo with poster)
  • RoboEye: a Wheelchair Plugin for Mobility Enhancement, Malvina Leuci, Luca Maule, Alberto Fornaser, Mariolino De Cecco (Università degli Studi di Trento),  Alfredo Armanini (Xtensa s.r.l.), Nicola Conci, Francesco De Natale (Università degli Studi di Trento)
  • Splicebuster: a New Blind Image Splicing Detector, Davide Cozzolino, Giovanni Poggi and Luisa Verdoliva (Università degli Studi di Napoli Federico II)

Posters:

  • A Game-Theoretic Framework for Optimum Decision Fusion in the Presence of Byzantines, Andrea Abrardo, Mauro Barni, Kassem Kallas and Benedetta Tondi (Università degli Studi di Siena)
  • A Visual Sensor Network for Parking Lot Occupancy Detection in Smart Cities, Luca Baroffio, Luca Bondi, Matteo Cesana, Alessandro Redondi and Marco Tagliasacchi (Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano)
  • Deep Convolutional Neural Networks for Pedestrian Detection, Denis Tomè, Federico Monti, Luca Baroffio, Luca Bondi, Marco Tagliasacchi and Stefano Tubaro (Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano)
  • Exploiting Silhouette Descriptors and Synthetic Data for Hand Gesture Recognition, Ludovico Minto, Alvise Memo and Pietro Zanuttigh (Università di Padova)
  • Fetal Heart Rate Estimation from Compressed Sensed Abdominal Fetal ECG Recordings, Giulia Da Poian, Riccardo Bernardini and Roberto Rinaldo (Dipartimento di Ingegneria Elettrica, Gestionale e Meccanica, Università degli Studi di Udine)
  • Splitting and Merging Schemes for Color and Depth Segmentation Based on Nurbs Fitting, Giampaolo Pagnutti and Pietro Zanuttigh, Università degli Studi di Padova
  • Smartphone Fingerprint by Mixing Device Sensors Features, Irene Amerini (Media Integration and Communication Center (MICC), Università degli Studi di Firenze), Paolo Bestagini (Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano), Luca Bondi (Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano), Roberto Caldelli (Consorzio Nazionale Interuniversitario per le Telecomunicazioni (CNIT), Università degli Studi di Parma, Media Integration and Communication Center (MICC), Università degli Studi di Firenze), Matteo Casini (Media Integration and Communication Center (MICC), Università degli Studi di Firenze) and Stefano Tubaro (Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano)
  • Sparsity Estimation from Compressive Projections via Sparse Random Matrices, Chiara Ravazzi, Sophie M. Fosson, Tiziano Bianchi and Enrico Magli (Politecnico di Torino)


The Future of the Italian DSP/MMSP Community (Summer School, PRIN proposals, VQR, ASN)

Riccardo Leonardi (Università degli Studi di Brescia)

Time: 15:00 - 16:30


Invited Speakers

Yoram Bresler

Yoram Bresler (F) received the B.Sc. (cum laude) and M.Sc. degrees from the Technion, Israel Institute of Technology, (1974 and 1981 respectively); the Ph.D degree from Stanford University, (1986), all in Electrical Engineering. Since 1987, he has been on the faculty at the University of Illinois, Urbana-Champaign, where he is currently Professor, Department of Electrical and Computer Engineering and the Department of Bioengineering, and at the Coordinated Science Laboratory. In 2003, Dr. Bresler co-founded InstaRecon, Inc., based in Champaign, Illinois, to commercialize breakthrough technology for tomographic reconstruction developed in his academic research. He currently serves as the company’s President and Chief Technology Officer.

Dr. Bresler was elected IEEE Fellow in 1999 ”for contributions to computer-based imaging and sensor array processing,” and in 2010, Fellow of the American Institute for Medical and Biomedical Engineering (AIMBE), ”for pioneering contributions to fast tomographic reconstruction algorithms and fundamental contributions to sampling theory for fast dynamic imaging.” He holds 11 US patents and more than 20 international patents, and has received the IEEE SPS Best Paper Award (1988) and (1989). He is the recipient of a NSF Presidential Young Investigator Award (1991); the Technion (Israel Inst. of Technology) Fellowship (1995); and the Xerox Senior Award for Faculty Research (1998). He was named a University of Illinois Scholar (1999); appointed as Associate, the Center for Advanced Study of the University (2001-2002); and Faculty Fellow, the NCSA (National Center for Super Computing Applications) (2006).

Dr. Bresler has served as Associate Editor, IEEE Transactions on Signal Processing (1992-1993); on Editorial Boards, Machine Vision and Applications (1987-2006), and the SIAM Journal on Imaging Science (2007-2013); on the Senior Editorial Board, IEEE Journal on Selected Topics in Signal Processing (2006-2013); and he was Guest co-Editor, IEEE Transactions on Medical Imaging special issue on Compressed Sensing. He was a member, IEEE Image and Multidimensional Signal Processing Technical Committee (1993 – 1998) and the IEEE Bio Imaging and Signal Processing Technical Committee (2005-2009), and served on the IEEE SPS Awards Board (2003-2006).

Dr. Bresler’s interests are in multi-dimensional and statistical signal processing and their applications to inverse problems in imaging, and in particular compressed sensing, which he introduced with his students in the mid 90’s under the monikers of “spectrum-blind sampling,” and “image compression on the fly,” as well as computed tomography, magnetic resonance imaging, and learning-based signal processing. 

Sergio Barbarossa

Sergio Barbarossa is a Full Professor at the University of Rome "La Sapienza". He has held positions as a Research Engineer with Selenia SpA (1984-1986), with the Environmental Institute of Michigan (1988), as a Visiting Professor with the University of Virginia (1995 and 1997) and with the University of Minnesota (1999). He is the author of a research monograph titled ``Multiantenna Wireless Communication Systems''.
Since 2000 he has been deeply involved in European projects. He has been the scientific coordinator of the European projects on wireless sensor networks, femtocell networks, spaceborne radar remote sensing, and mobile cloud computing. Dr. Barbarossa is an IEEE Fellow and EURASIP Fellow. He received the 2010 EURASIP Technical Achievements Award for his contributions to synthetic aperture radar, sensor networks, and communication networks. He received the IEEE Best Paper Awards from the IEEE Signal Processing Society in 2000 and 2014. He has been an IEEE Distinguished Lecturer from the Signal Processing Society in 2012-2013. He coauthored papers that received the Best Student Paper Award at ICASSP 2006, SPAWC 2010, EUSIPCO 2011, and CAMSAP 2011. From 1997 until 2003, he has been a member of the IEEE Technical Committee for Signal Processing in Communications. He served as an Associate Editor for the IEEE Transactions on Signal Processing and for the IEEE Signal Processing Magazine.
He is currently an Associate Editor of the IEEE Transactions on Signal and Information Processing over Networks. He has been the General Chairman of the IEEE Workshop on Signal Processing Advances in Wireless Communications (SPAWC) 2003 and the Technical Co-Chair of SPAWC 2013. He has been the Guest Editor for Special Issues on the IEEE Journal on Selected Areas of Communications, IEEE Signal Processing Magazine, IEEE Journal of Selected Topics in Signal Processing, EURASIP Journal of Applied Signal Processing, EURASIP Journal on Wireless Communications and Networking. His current research interests lie in the area of signal processing over graphs and hypergraphs, adaptation and learning over networks, mobile edge computing, and distributed optimization algorithms.

Social Events

Social activities

The morning of Monday, 25th January is reserved for winter activities!
  • For those interested in skiing: The Alleghe ski resort (skicivetta) can be accessed with the cable car just in front of the workshop location. For more details have a look at  http://www.skicivetta.com or http://www.dolomitisuperski.com .
  • For non-skiers: we are organizing a snow-shoe excursion ("ciaspolata"). The cost will be around  € 20 including shoes (€ 10 without the shoes). More details will be available soon.
  • For the lazy ones: enjoy the beautiful scenery, the lake, the food and the wine!

Social dinner

The social dinner will take place on Monday evening. The dinner will be "on the slopes" at the Grande Baita Civetta . The restaurant is at Piani di Pezzé (close to the upper station of the first section of the cable car from Alleghe). The location will be reached by car up to Piani di Pezzé and then with a small snowcat transfer. More details will be given at the workshop.