Cosimo DISTANTE

Cosimo DISTANTE

Docente a contratto

Image Processing

Area di competenza:

Computer Vision and Pattern Recognition

tel 0832 1975300

Curriculum Vitae

 

Cosimo Distante PhD – Institute of Applied Sciences and Intelligent Systems - ISASI CNR

Master in Computer Science at the University Bari, and PhD in Engineering at University of Salento (Italy). His main expertise are in the field of Artificial Intelligent, in particular Computer Vision and Pattern Recognition, Image Processing, Machine Learning and Robotics. Particular emphasis is devoted to the development of Deep Learning algorithms. He has been a visiting researcher at the Computer Science Department of the University of Massachusetts (Amherst, MA) where has carried out research activities under robotics, artificial neural networks and vision. Dr. Distante joined the Faculty of the University of Massachusetts as a Teaching Assistant, for the "Artificial Intelligence" 683 class of the Master in Sciences 1998. He received the Ph.D. in Material Engineering from the University of Salento. Dr. Distante joined in 2001 the Italian National Research Council of Italy CNR, focusing his research activity in the context of robot learning, visual grasping and pattern recognition applied to array of chemical sensors. Dr. Distante has been awarded with several fellowships, among them the CNR-NATO in which he has collaborated with the University of Manchester (Prof. Krishna Persaud) for the study of pattern recognition techniques applied to the context of gas sensor arrays for parameter drift reduction. Since 2003 he is Contract Professor for the courses of Pattern Recognition and Image Processing in Computer Engineering at the University of the Salento IT. Since 2010 he started a technology transfer activity by establishing a startup spinoff company named Taggalo.

In 2011, Dr. Distante has been awarded with the national innovation Prize working capital PNI-Cube TelecomItalia with the Taggalo project. He won the ChallengUP (Cisco, Intel and Deutch Telekom acceleration program) contest in 2015 with Taggalo. He served as general chair of several established conferences, among them the IEEE AVSS 2017 that focus on videosurveillance, the Video analytics for Audience mearuserement that focus on computer vision tech. for retail and digital signage. He is coordinating several joint research lab between CNR and Leonardo-Finmeccanica for security technology that focus on face recognition in the wild and videosurveillance, and between CNR and a pool of SMEs for the health sector. He is the unit manager of the Institute of Applied sciences and Intelligent system of the CNR where leads a group of researchers, PhD students, coordinates two industrial joint labs (security and health).

He is expert member in innovation for the Ministry of Education, Research and University (MIUR), as well as the Ministry of economic development (MISE) and a few regional agencies for technology and innovation.

Image Processing

II Semester

Scarica curriculum vitae

Didattica

A.A. 2020/2021

COMPUTER VISION

Degree course COMPUTER ENGINEERING

Course type Laurea Magistrale

Language INGLESE

Credits 9.0

Teaching hours Ore totali di attività frontale: 81.0

Year taught 2020/2021

For matriculated on 2020/2021

Course year 1

Structure DIPARTIMENTO DI INGEGNERIA DELL'INNOVAZIONE

Subject matter PERCORSO COMUNE

Location Lecce

A.A. 2019/2020

COMPUTER VISION

Degree course COMPUTER ENGINEERING

Course type Laurea Magistrale

Language INGLESE

Credits 9.0

Teaching hours Ore totali di attività frontale: 81.0

Year taught 2019/2020

For matriculated on 2019/2020

Course year 1

Structure DIPARTIMENTO DI INGEGNERIA DELL'INNOVAZIONE

Subject matter PERCORSO COMUNE

Location Lecce

A.A. 2018/2019

IMAGE PROCESSING

Degree course COMPUTER ENGINEERING

Course type Laurea Magistrale

Language INGLESE

Credits 9.0

Teaching hours Ore totali di attività frontale: 81.0

Year taught 2018/2019

For matriculated on 2018/2019

Course year 1

Structure DIPARTIMENTO DI INGEGNERIA DELL'INNOVAZIONE

Subject matter PERCORSO COMUNE

Location Lecce

A.A. 2017/2018

IMAGE PROCESSING

Degree course COMPUTER ENGINEERING

Course type Laurea Magistrale

Language INGLESE

Credits 9.0

Teaching hours Ore totali di attività frontale: 0.0

Year taught 2017/2018

For matriculated on 2017/2018

Course year 1

Structure DIPARTIMENTO DI INGEGNERIA DELL'INNOVAZIONE

Subject matter PERCORSO COMUNE

Location Lecce

Torna all'elenco
COMPUTER VISION

Degree course COMPUTER ENGINEERING

Subject area ING-INF/03

Course type Laurea Magistrale

Credits 9.0

Teaching hours Ore totali di attività frontale: 81.0

For matriculated on 2020/2021

Year taught 2020/2021

Course year 1

Semestre Primo Semestre (dal 22/09/2020 al 18/12/2020)

Language INGLESE

Subject matter PERCORSO COMUNE (999)

Location Lecce

No prior experience with computer vision is assumed, although previous knowledge of visual computing or signal processing will be helpful. The following skills are necessary for this class:

  • Math: Linear algebra, vector calculus, and probability. Linear algebra is the most important.
  • Data structures: Students will write code that represents images as feature and geometric constructions.
  • Programming: A good working knowledge. All lecture code and project starter code will be Python, and Pytorch for Deep Learning, but student familiar with other frameworks such as tensorflow is ok. 

Computer Vision today is everywhere in our society and images have become pervasive, with applications in several sectors; just to mention some in: apps, drones, healthcare and precision medicine, precision agricolture, searching, understanding, control in robotics and self-driving cars.

The course introduces the basics of image formation, reconstruction and inferring motion models, as well as camera calibration theory and practice.

Recent developments in neural networks (Deep Learning) have considerably boosted the performance of the visual recognition systems in tasks such as: classification, localisation, detection, segmentation etc. Students will learn the building blocks of a general convolutional neural network, the way how it is trained and optimized, how to prepare a dataset and how to measure the final performance.

Upon completion of this course, students will:

  1. Be familiar with both the theoretical and practical aspects of computing with images;
  2. Have described the foundation of image formation, measurement, and analysis;
  3. Have implemented common methods for robust image matching and alignment;
  4. Understand the geometric relationships between 2D images and the 3D world;
  5. Have gained exposure to object and scene recognition and categorization from images;
  6. Grasp the principles of state-of-the-art deep neural networks; and
  7. Developed the practical skills necessary to build computer vision applications.

Teaching is based on theoretical and practical lectures. The student will write in python algorithms taught in class

Oral session. The student will explain the developed project and shall answer two or more questions regarding theoretical aspects of the studied topics

The student must develop a project by choosing a practical simple application with some algorithms done during the course. The choice is at total disposal of the student, as well as the fact of developing it in group os solo. In group setting the students must proof their own activities developed in the common project application.

The final examination is based on oral assessment of the topics covered during lectures.

For the LAB practice, students may use for the deep learning development the Google Colab or Cloud Platform.

Introduction to Computer Vision

Image Formation

2D and 3D geometric primitives - Projections

image enhancement

LAB Introduction to Python and Operations with images

Color perception, color spaces and processing

Image Filtering

image pyramids and blending

Local feature detector

LAB SIFT with MatLab Find image rotation & scale SURF Object Detection

Image Alignment I- warping, homography estimation direct linear transform

Image Alignment II- robust motion estimation with Ransac - perspective n point problem. Registration examples: face recognition, medical imaging

Camera Calibration - distortion models and compensations - linear methods for camera parameters. Calibration with a checkerboard

LAB Mosaicking with SURF Face Detection and Tracking (Nose Skin) Face Detection and Tracking withKanade - Lucas - Tomasi feature tracker

Motion Analysis and background modelling, application to intelligent videosurveillance 

Multiview geometry - Epipolar geometry, position error estimation, stereo rig, Essential matrix estimation, rectification, Reconstruction, correspondense problem, weak calibration and ransac estimation of fundamental matrix

LAB - Camera calibration

Image Classification - Key nearest neighbor, linear classifiers

Image Classification - loss functions, optimization with stochastic gradient descent

LAB - Stereo calibration and reconstruction

Image Classification - backpropagation and neural networks, computational graphs and gradient estimation

Image Classification - Convolutional Neural Network architecture

Image Classification - CNN activation functions, data preprocessing, weight normalization, batch normalization, monitoring the learning process, hyperparameter optimization, Regularization (Dropout, drop connect, fractional pooling, cotout, mixup)

Image Classification - CNN activation functions, data preprocessing, weight normalization, batch normalization, monitoring the learning process, hyperparameter optimization, Regularization (Dropout, drop connect, fractional pooling, cotout, mixup)

Image Classification - Object detection and Image segmentation with Convolutional neural networks. Introduction to auto-encoders

LAB - Introduction to Pytorch framework

LAB - Deep learning applications to object detection (Yolo and Faster R-CNN)

LAB - Deep Learning application to segmentation with mark R-CNN

 

There is no requirement to buy a book. The goal of the course is to be self contained, but sections from the following textbooks will be suggested for more formalization and information.

The primary course text will be Rick Szeliski’s draft Computer Vision: Algorithms and Applications; we will use an online copy of the June 19th draft.  A copy and link will be provided in website.  The secondary  text is Forsyth and Ponce, Computer Vision: A Modern Approach (new Edition coming out in 2020)

Deep Learning, MIT Press book, Ian Goodfellow and Yoshua Bengio and Aaron Courville

 

 

COMPUTER VISION (ING-INF/03)
COMPUTER VISION

Degree course COMPUTER ENGINEERING

Subject area ING-INF/03

Course type Laurea Magistrale

Credits 9.0

Teaching hours Ore totali di attività frontale: 81.0

For matriculated on 2019/2020

Year taught 2019/2020

Course year 1

Semestre Primo Semestre (dal 23/09/2019 al 20/12/2019)

Language INGLESE

Subject matter PERCORSO COMUNE (999)

Location Lecce

No prior experience with computer vision is assumed, although previous knowledge of visual computing or signal processing will be helpful. The following skills are necessary for this class:

  • Math: Linear algebra, vector calculus, and probability. Linear algebra is the most important.
  • Data structures: Students will write code that represents images as feature and geometric constructions.
  • Programming: A good working knowledge. All lecture code and project starter code will be Python, and Pytorch for Deep Learning, but student familiar with other frameworks such as tensorflow is ok. 

Computer Vision today is everywhere in our society and images have become pervasive, with applications in several sectors; just to mention some in: apps, drones, healthcare and precision medicine, precision agricolture, searching, understanding, control in robotics and self-driving cars.

The course introduces the basics of image formation, reconstruction and inferring motion models, as well as camera calibration theory and practice.

Recent developments in neural networks (Deep Learning) have considerably boosted the performance of the visual recognition systems in tasks such as: classification, localisation, detection, segmentation etc. Students will learn the building blocks of a general convolutional neural network, the way how it is trained and optimized, how to prepare a dataset and how to measure the final performance.

Upon completion of this course, students will:

  1. Be familiar with both the theoretical and practical aspects of computing with images;
  2. Have described the foundation of image formation, measurement, and analysis;
  3. Have implemented common methods for robust image matching and alignment;
  4. Understand the geometric relationships between 2D images and the 3D world;
  5. Have gained exposure to object and scene recognition and categorization from images;
  6. Grasp the principles of state-of-the-art deep neural networks; and
  7. Developed the practical skills necessary to build computer vision applications.

Teaching is based on theoretical and practical lectures. The student will write in python algorithms taught in class

Oral session. The student will explain the developed project and shall answer two or more questions regarding theoretical aspects of the studied topics

The student must develop a project by choosing a practical simple application with some algorithms done during the course. The choice is at total disposal of the student, as well as the fact of developing it in group os solo. In group setting the students must proof their own activities developed in the common project application.

The final examination is based on oral assessment of the topics covered during lectures.

For the LAB practice, students may use for the deep learning development the Google Colab or Cloud Platform.

Introduction to Computer Vision

Image Formation

2D and 3D geometric primitives - Projections

image enhancement

LAB Introduction to Python and Operations with images

Color perception, color spaces and processing

Image Filtering

image pyramids and blending

Local feature detector

LAB SIFT with MatLab Find image rotation & scale SURF Object Detection

Image Alignment I- warping, homography estimation direct linear transform

Image Alignment II- robust motion estimation with Ransac - perspective n point problem. Registration examples: face recognition, medical imaging

Camera Calibration - distortion models and compensations - linear methods for camera parameters. Calibration with a checkerboard

LAB Mosaicking with SURF Face Detection and Tracking (Nose Skin) Face Detection and Tracking withKanade - Lucas - Tomasi feature tracker

Motion Analysis and background modelling, application to intelligent videosurveillance 

Multiview geometry - Epipolar geometry, position error estimation, stereo rig, Essential matrix estimation, rectification, Reconstruction, correspondense problem, weak calibration and ransac estimation of fundamental matrix

LAB - Camera calibration

Image Classification - Key nearest neighbor, linear classifiers

Image Classification - loss functions, optimization with stochastic gradient descent

LAB - Stereo calibration and reconstruction

Image Classification - backpropagation and neural networks, computational graphs and gradient estimation

Image Classification - Convolutional Neural Network architecture

Image Classification - CNN activation functions, data preprocessing, weight normalization, batch normalization, monitoring the learning process, hyperparameter optimization, Regularization (Dropout, drop connect, fractional pooling, cotout, mixup)

Image Classification - CNN activation functions, data preprocessing, weight normalization, batch normalization, monitoring the learning process, hyperparameter optimization, Regularization (Dropout, drop connect, fractional pooling, cotout, mixup)

Image Classification - Object detection and Image segmentation with Convolutional neural networks. Introduction to auto-encoders

LAB - Introduction to Pytorch framework

LAB - Deep learning applications to object detection (Yolo and Faster R-CNN)

LAB - Deep Learning application to segmentation with mark R-CNN

 

There is no requirement to buy a book. The goal of the course is to be self contained, but sections from the following textbooks will be suggested for more formalization and information.

The primary course text will be Rick Szeliski’s draft Computer Vision: Algorithms and Applications; we will use an online copy of the June 19th draft.  A copy and link will be provided in website.  The secondary  text is Forsyth and Ponce, Computer Vision: A Modern Approach (new Edition coming out in 2020)

Deep Learning, MIT Press book, Ian Goodfellow and Yoshua Bengio and Aaron Courville

 

 

COMPUTER VISION (ING-INF/03)
IMAGE PROCESSING

Degree course COMPUTER ENGINEERING

Subject area ING-INF/03

Course type Laurea Magistrale

Credits 9.0

Teaching hours Ore totali di attività frontale: 81.0

For matriculated on 2018/2019

Year taught 2018/2019

Course year 1

Semestre Secondo Semestre (dal 04/03/2019 al 04/06/2019)

Language INGLESE

Subject matter PERCORSO COMUNE (999)

Location Lecce

IMAGE PROCESSING (ING-INF/03)
IMAGE PROCESSING

Degree course COMPUTER ENGINEERING

Subject area ING-INF/03

Course type Laurea Magistrale

Credits 9.0

Teaching hours Ore totali di attività frontale: 0.0

For matriculated on 2017/2018

Year taught 2017/2018

Course year 1

Semestre Secondo Semestre (dal 01/03/2018 al 01/06/2018)

Language INGLESE

Subject matter PERCORSO COMUNE (999)

Location Lecce

The goal of image processing class is to provide the fundamentals of developing an intelligent machine vision system. The goal is to study and analyse images and videos to understand their content and derive 3D information. Problems in this field concern the identification of 3D shapes of an acquired scene, to determine how objects move, and recognize objects through the analysis of still images or a sequence of them (ie through static and / or time-varying information). The course provides an introduction to classical image processing techniques and end up to introduce the Deep Learning methodologies that are nowadays at the basis of all the disrupting innovations in several sectors: self-driving cars, security for face recognition and behaviour understanding, precision medicine and agricolture etc

at the end of the course the student will be able to:
be familiar with the theoretical and practical aspects of image processing; have acquired the basics of the image formation process and understand the relationships between the 2D and 3D world; have acquired the essential ingredients to develop a processing pipeline to locate, recognize and track objects of interest.
Having acquired the basic principles of Deep Neural Networks (Deep Learning) and transfer learning in order to build intelligent vision systems

Introduzione ai sistemi di visione artificiale (2 ore); Formazione dell’immagine (3 ore); Geometria proiettiva 2D e 3D (3 ore); Miglioramento della qualità delle immagini (2 ore); analisi delle immagini a colori (2 ore); Filtraggio nello spazio e nel dominio delle frequenze (4 ore); Piramidi Gaussiane e Laplaciane (3 ore); Local feature detector (4 ore); Allineamento (4 ore); Segmentazione (3 ore); analisi della tessitura (2 ore); analisi del movimento (4 ore); structure from motion (2 ore); multi-view geometry (2); Riconoscimento automatico (2) Deep Learning (8 ore); Tracking (2 ore).

[1] Richard Szeliski, Computer Vision: Algorithms and Applications, Springer 2010.

[2] Deep Learning, by Goodfellow, Bengio, and Courville.

[2] Dictionary of Computer Vision and Image Processing, by Fisher et al. Note: Full text is available in 'Online Resources' section.

IMAGE PROCESSING (ING-INF/03)
IMAGE PROCESSING

Corso di laurea COMPUTER ENGINEERING

Settore Scientifico Disciplinare ING-INF/03

Tipo corso di studio Laurea Magistrale

Crediti 9.0

Ripartizione oraria Ore totali di attività frontale: 0.0

Per immatricolati nel 2016/2017

Anno accademico di erogazione 2016/2017

Anno di corso 1

Semestre Secondo Semestre (dal 01/03/2017 al 02/06/2017)

Lingua

Percorso PERCORSO COMUNE (999)

Sede Lecce - Università degli Studi

IMAGE PROCESSING (ING-INF/03)
IMAGE PROCESSING

Corso di laurea COMPUTER ENGINEERING

Settore Scientifico Disciplinare ING-INF/03

Tipo corso di studio Laurea Magistrale

Crediti 9.0

Ripartizione oraria Ore totali di attività frontale: 0.0

Per immatricolati nel 2015/2016

Anno accademico di erogazione 2015/2016

Anno di corso 1

Semestre Primo Semestre (dal 21/09/2015 al 18/12/2015)

Lingua

Percorso PERCORSO COMUNE (999)

Sede Lecce - Università degli Studi

IMAGE PROCESSING (ING-INF/03)
IMAGE PROCESSING

Corso di laurea COMPUTER ENGINEERING

Settore Scientifico Disciplinare ING-INF/03

Tipo corso di studio Laurea Magistrale

Crediti 9.0

Ripartizione oraria Ore totali di attività frontale: 0.0

Per immatricolati nel 2014/2015

Anno accademico di erogazione 2014/2015

Anno di corso 1

Semestre Primo Semestre (dal 29/09/2014 al 13/01/2015)

Lingua

Percorso PERCORSO COMUNE (999)

Sede Lecce - Università degli Studi

IMAGE PROCESSING (ING-INF/03)
IMAGE PROCESSING

Corso di laurea COMPUTER ENGINEERING

Settore Scientifico Disciplinare ING-INF/03

Tipo corso di studio Laurea Magistrale

Crediti 9.0

Ripartizione oraria Ore totali di attività frontale: 0.0

Per immatricolati nel 2013/2014

Anno accademico di erogazione 2013/2014

Anno di corso 1

Semestre Primo Semestre (dal 30/09/2013 al 21/12/2013)

Lingua

Percorso PERCORSO COMUNE (999)

Sede Lecce - Università degli Studi

IMAGE PROCESSING (ING-INF/03)

Pubblicazioni