Course timetables for the first semester of the academic year 2014/2015 are available here.

 

BS courses

Teacher: Francesco De Natale, Lorenzo Bruzzone, Andrea Rosani, Claudia Paris.
Date: second semester.
Degree: BS degree in Electronics and Telecommunication Engineering.
The course is taught in italian.

Go to ESSE3 services.

Description

The course analyses the fundamentals of a numerical transmission system. Starting from the description of the A/D and D/A conversion processes, the main baseband and passband transmission systems are faced. The course includes exercise and laboratory sessions.

Topics

Signal sampling: introduction the A/D-D/A conversion, ideal sampling, Nyquist-Shannon sampling theorem, Sampled signal reconstruction (time-frequency analysis), interpolating filters, time-limited signals and spectral resolution, real sampling (sample and hold, chopper).

Pulse code modulation (PCM): basic concepts, uniform quantization, PCM encoder and decoder, quantization noise, non-uniform quantization.

Bandbase numerical transmission: introduction to digital transmitters, pulse amplitude modulation (PAM), unlimited bandwidthand band-limited channels, PAM for unlimited bandwidth and narrowband channels, line code, PAM with memory, intersymbol interference, 2nd Nyquist condition, equalization, eye diagram, PAM spectrum.

Multiplexing systems: TDM systems, synchronous/asynchronous systems.

Passband numerical transmission: bandpass and lowpass representation of a modulated signal, evaluation of digital modulation performance, modulation methods (OOK, ASK, QAM, PRK, FSK, Sunde FSK, CPFSK, MSK), amplitude phase mixed modulation (APK), spectrum, spectral efficiency of modulation techniques, error probability, performance comparison of digital modulation methods, comparison of digital and analog transmission techniques.

 References

A. B. Carlson, Communication Systems, McGraw-Hill, 1986
S. Haykin, Digital Communications, John Wiley, 1988.
J.G. Proakis, Digital Communications, McGraw-Hill, terza edizione, 1995

Teacher: Francesco De Natale, Andrea Rosani.
Date: first semester.
Degree: BS degree in Electronics and Telecommunication Engineering.
The course is taught in italian.

Go to ESSE3 services.

Description

The course analyses the main aspects of a processing, storing and trasmission system of multimedia signals.
In the first part of the course, fundamentals of signal processing in the numeric domain are studied. Then, numerical images and related problems are considered, like acquisition, A/D conversion, color representation. Moreover, image processing techniques based on histogram, calibration and geometrical correction are presented, as well as segmentation and image characterization methods. Finally, the problem of image storage is analyzed, introducing the most common methodologies and standards (JPEG, JPEG2000).

Topics

Introduction to signals and numerical processing systems.
Introduction to digital imaging.
Acquisition, A/D conversion, canonical form, chromatic spaces, image formats.
Pre-processing, histograms and histogram smoothing techniques, calibration, geometric corrections, digital filters.
Problems in image acquisition (noise, distortion, etc.), linear filters, 2D convolution, 2D Fourier. transform, frequency representation.
Image segmentation, gradient edge-detection, Sobel algorithm.
Region extraction (threshold method, clustering, split-and-merge,region growing).
Image characterization with features.
Basic concepts of image compression.
Vectorial quantization, subband decomposition, piramidal decomposition.
Transform coding, JPEG standard and advances.
Standard system of audio-visual communication.

References

A.K. Jain, Fundamentals of Digital Image Processing, Prentice-Hall.
A. N. Netravali, B. G. Haskell, Digital Pictures – Representation, Compression and Standards, Plenum Press.

Teacher: Francesco De Natale, Andrea Rosani.
Date: first semester.
Degree: BS degree in Electronic and Telecommunication Engineering.
The course is taught in italian.

Go to ESSE3 services.

Description

The course analyses the main aspects of a processing, storing and trasmission system of multimedia signals.
In the first part of the course, fundamentals of signal processing in the numeric domain are studied. Then, numerical images and related problems are considered, like acquisition, A/D conversion, color representation. Moreover, image processing techniques based on histogram, calibration and geometrical correction are presented, as well as segmentation and image characterization methods. Finally, the problem of image storage is analyzed, introducing the most common methodologies and standards (JPEG, JPEG2000).

Topics

Introduction to signals and numerical processing systems.
Introduction to digital imaging.
Acquisition, A/D conversion, canonical form, chromatic spaces, image formats.
Pre-processing, histograms and histogram smoothing techniques, calibration, geometric corrections, digital filters.
Problems in image acquisition (noise, distortion, etc.), linear filters, 2D convolution, 2D Fourier. transform, frequency representation.
Image segmentation, gradient edge-detection, Sobel algorithm.
Region extraction (threshold method, clustering, split-and-merge,region growing).
Image characterization with features.
Basic concepts of image compression.
Vectorial quantization, subband decomposition, piramidal decomposition.
Transform coding, JPEG standard and advances.
Standard system of audio-visual communication.

References

A.K. Jain, , Fundamentals of Digital Image Processing, Prentice-Hall.
A. N. Netravali, B. G. Haskell, Digital Pictures – Representation, Compression and Standards, Plenum Press.

MS courses

Teacher: Nicola Conci, Habib Ullah.
Date: first semester.
Degree: MS degree in Telecommunication Engineering.
Go to ESSE3 services.

Description

The course aims at providing the student an overview of analysis methods in the field of Computer Vision. Starting from the main concepts in image/video processing, the course will be focused on problems of motion modeling and detection, tracking, object recognition, both considering monocular and multi-view systems.

Topics

  • Introduction to the concepts of digital image and video
  • Model definition (camera, illumination, object, scene)
  • Motion parametrization (camera, object)
  • Different kinds of projection, resolution, focus
  • 2D and 3D rigid transformation
  • Motion identification (background subtraction, frame difference)
  • Video segmentation
  • Stereo- and Multi-view
  • Tracking: methods based on color histogram, Bayesian tracking (Kalman filter, particle filters)
  • Classification in Computer Vision
  • Introduction to Information Retrieval: Computer Vision techniques for content extraction in images and video, based on color matching, shape matching, etc.
  • Matlab and OpenCV for image/video processing

References

L. G. Shapiro, Computer Vision
D. Forsyth, Computer Vision: a modern approach
A. Bovik, The essential guide to video processing
Hartley, Zisserman, Multiple View Geometry.

Teacher: Giulia Boato, Valentina Conotter.
Date: first semester.
Degree: MS degree in Telecommunication Engineering.
Go to ESSE3 services.

Description

The easy and low-cost access to enormous amounts of multimedia content leads nowaday to the problem of protecting and authenticating data. The course aims at exploring the main methods used to make secure the access to multimedia data, by means of data hiding techniques and automatic detection of manipulations in multimedia objects.

Topics

After an introduction on concepts and models of Digital Rights Management for multimedia data protection, the fields of digital watermarking and multimedia forensics are presented. Starting from a general analysis of the main concepts, the course will face the description and evaluation of specific techniques applied to multimedia objects.

Introduction to the problem of data protection

Digital Watermarking:

– history
– applications and models
– method classification and evaluation
– robustness and security
– watermarking algorithms for digital images.

Digital image forensics

– analysis and detection of manipulations in digital images
– statistical techniques
– geometric techniques
– image source identification

Teacher: Nicola Conci, Daniele Miorandi, Krishna Konda Reddy.
Date: second semester.
Degree: MS in Telecommunication Engineering.
Go to ESSE3 services.

Description

The course aims at providing the student an overview on communication and multimedia networking systems. The course includes multimedia processing concepts, architectures and network protocols supporting multimedia traffic on IP networks.

Topics

  • Main concepts of audio/video coding (hybrid predictive models)
  • Multimedia coding standard (i.e., MPEG-2/4, H.264 etc.)
  • Architectures for networks supporting multimedia contents
  • Network protocols for mnultimedia (e.g., RTP, RTSP, SIP, HTTP streaming)
  • Multimedia on IP networks: problems e solutions
  • Multimedia on wireless networks: problems and solutions
  • Case-study: Skype, YouTube, Akamai
  • Value chains

References

Suggested bibliography:
M. Tekalp, “Digital Video Processing”
A. Bovik, “The essential guide to Video Processing”
A. B. Johnston, “SIP: understanding the Session Initiation Protocol”
C. Perkins, “RTP: Audio and Video for the Internet”
J. Crowcroft, M. Handley, I. Wakeman, “Internetworking Multimedia”
J.D. Gibson (Ed), “Multimedia communications: directions and innovation”

 

Courses within the ICT Doctoral School Program

External lecturer: Ivan Laptev, INRIA Paris – Rocquencort.
Internal proponents: Nicu Sebe, Francesco De Natale.
Date: 7-11 July 2014.

Description

Automated  object  recognition – and  more generally  scene  analysis – from  photographs  and videos  is  the grand  challenge  of  computer  vision.  This  course  presents  the image,  object,  and scene models, as well as the methods and algorithms, used today to address this challenge.

Topics

1. Instance-level recognition I. – Camera geometry, Local invariant features
2. Instance-level recognition II. – Correspondence, efficient visual search
3. Instance-level recognition III. – Very large scale image indexing
4. Category-level recognition I. – Bag-of-feature models
5. Category-level recognition II. – Convolutional neural networks
6. Category-level localization I. – Deformable part models
7. Category-level localization II. – Efficient fitting of pictorial structures; Human pose estimation
8. Motion and human actions I. – Action classification with local features 9. Motion and human actions II. – Joint models for actions, objects and scenes
10. Student project presentations

More info here.