Technical Program

This is the final GTTI MMSP 2017 schedule. Click on a specific event to explore the details and to add it to your personal agenda.




Monday, January 30th

Keynote Speech: Multimodal Spatio-Temporal Signal Processing and Audio-Visual Perception

Petros Maragos
Time: 14:00 - 15:30
Abstract: In this talk we will present an overview of ideas, methods and research results in multimodal spatio-temporal sensory processing with emphasis on audio-visual signal processing and fusion as applied to problems of attention, inversion and recognition. We shall begin with a brief synopsis of important findings from audio-visual (A-V) perception. Then we shall outline efficient signal processing front-ends and fusion schemes for the problems of A-V speech recognition as well as A-V speech inversion to geometry. Spatio-temporal processing ideas are also applied to multi-sensor a smart home environment. Afterwards, emphasis will be given to problems of attention where we will present improved computational saliency models for audio and visual salient event detection, followed by multimodal saliency estimation; this leads to movie video summarization based on audio, visual, and text modalities. Finally, we will outline the application of some of the above ideas and methods to audio-gestural (visual gesture and spoken command) recognition for problems of human-robot interaction.
More information and related papers can be found in, and

Session 1: Student activities and opportunities with the Signal Processing Society

Session chair: Patrizio Campisi (Università di Roma Tre)
Time: 15:30 - 16:00

Demo and poster session

Session chair: Lucio Marcenaro (Università degli Studi di Genova)
Time: 16:30 - 19:00


Tuesday, January 31th

Session 2: Multimedia Forensics

Session chair: Mauro Barni (University of Siena) and Alessandro Piva (University of Florence)
Time: 9:00 - 10:30
  1. Academic research towards real forensic scenarios: methodologies for investigating multimedia content, Marco Fontani, Ph.D., FORLAB Multimedia Forensics Laboratory, PIN s.c.r.l. Polo Universitario Città di Prato (University of Florence)
  2. Evtech-EC3 Platform – EU benchmark platform for assessing tools contributing to identification of authors and victims of Child Abuses online, Pasquale Ferrara, Ph.D., DIR. E - Space, Security and Migration, Joint Research Centre, European Commission
  3. Forensic Image and Video Enhancement: case studies and perspectives, Prof. Sebastiano Battiato, Dipartimento di Matematica ed Informatica, Università di Catania
  4. Near-Duplicate Video Detection Exploiting Noise Residual Traces, Prof. Stefano Tubaro (and Silvia Lameri, Luca Bondi, Paolo Bestagini), Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano
  5. Blind image forgery localization: an anomaly detection perspective, Prof. Luisa Verdoliva (and Davide Cozzolino, Giovanni Poggi), Dipartimento di Ingegneria Elettronica e Telecomunicazioni, Università Federico II di Napoli

Session 3: Signal Processing meets Deep Learning

Session chair: Giovanni Poggi (Università degli Studi di Napoli Federico II)
Time: 10:45 - 12:30
  1. Introduction, Prof. Giovanni Poggi, Università degli Studi di Napoli Federico II
  2. Recasting handcrafted features as CNNs - an application to image forgery detection, Prof. Giovanni Poggi, Università degli Studi di Napoli Federico II
  3. Deep learning using Bayesian Normal Graphs, Prof. Francesco A. N. Palmieri, Università della Campania "Luigi Vanvitelli" (ex SUN)
  4. Segmentation and Semantic Labeling Driven by Deep Learning and Surface Fitting, Prof. Pietro Zanuttigh, Dipartimento di Ingegneria dell’Informazione, Università di Padova
  5. Deep Convolutional Neural Networks for Classification of Mild Cognitive Impaired and Alzheimer’s Disease Patients from Scalp EEG Recordings, Cosimo Ieracitano, DICEAM - Dipartimento di Ingegneria Civile, dell'Energia, dell'Ambiente e dei Materiali, Università Mediterranea di Reggio Calabria
  6. Deep learning in new era of digital microbiology imaging, Prof. Alberto Signoroni, Dipartimento di Ingegneria dell'Informazione, Università di Brescia

Session 4: Sharing thoughts for DSP research/education towards AI - are these essential components for Industry 4.0?

Session chair: Riccardo Leonardi (Università di Brescia)
Time: 14:00 - 15:30