COMPUTER VISION E LABORATORIO

Module LABORATORIO

Knowledge and understanding: The objective of the course is to provide knowledge that enables the student to understand the theoretical and physical mechanisms underlying an artificial vision system.

Applying knowledge and understanding: The student will acquire the skills necessary to fully configure a system, choosing the optimal hardware/software configuration. In this respect, part of the course will consist of laboratory lessons, with practical examples of configuration and use of dedicated libraries, including in mobile environments.

Making judgements: Through real field simulation sessions, the student will be able to independently develop solutions capable of solving the main practical and design issues of a vision system (parameter setting, real-time processing, etc.).

Communication skills: The student will acquire the necessary communication skills and expressive appropriateness in the use of technical language in the general context of information systems and computer vision in particular.

Learning skills: The course aims to provide the student with the theoretical and practical methodologies necessary to independently face and solve new problems that may arise during work activities. For this purpose, several topics will be addressed during lessons, involving the student in the search for possible solutions to real problems.

The course aims to provide a broad and in-depth overview of the theories and techniques of computer vision, presenting their main practical and scientific applications. In the first part, students will be guided through the study of Image Formation Models, Calibration, and Stereoscopy, followed by exploration of filters and characteristics of Event Cameras, such as edge detection, texture analysis, and Laplacian pyramid construction. Techniques for detecting interest points (like Corner Detection, Harris, etc.), analysis of the SIFT model with its theoretical and practical principles, and visual segmentation methods will also be covered.

The second part of the course is dedicated to vision models applied to concrete cases, such as face recognition, image and video reconstruction and restoration, and the study of Visual Foundation Models, with particular attention to their implications and emerging applications. Each topic will be discussed in detail, covering both traditional approaches and the latest developments based on new-generation neural architectures such as GANs and diffusion models

1. E. Trucco, A. Verri, “Introductory Techniques for 3-D Computer Vision”, Prentice Hall, 1998
2. G. Bradski, A. Kaehler, “Learning OpenCV Computer Vision with the OpenCV Library” O'Reilly Media, 2008
   
3. Mubarak Shah, "Fundamentals of Computer Vision" (pdf), 1997
   
4. R. Hartley and A. Zisserman, “Multiple View Geometry in Computer Vision”, 2004
   
5. D. A. Forsyth, J. Ponce, “Computer Vision a Modern Approach”, Prentice Hall PTR, 2002
   
6. Richard Szeliski, Computer Vision: Algorithms and Application, Springer 2010
   
7. Fondamenti di Image Processing - Guida teorico/pratica per l’elaborazione e la codifica di immagini digitali – ISBN: 88-88659-49-8 - EdiArgo 2006
   
8. Elaborazione delle Immagini Digitali - R.C. Gonzales, R.E. Woods – Pearson Italia 2008

Oral examination and an individual project to be agreed upon with the instructor.

The oral exam is structured so that each student is given a grade according to the following scheme:

• Not approved: the student has not acquired the basic concepts and cannot answer at least 60% of the questions nor solve the exercises.
• 18–23: the student demonstrates minimal mastery of basic concepts; their ability to connect topics is modest and they solve simple exercises.
• 24–27: the student demonstrates good mastery of course content; their ability to connect topics is good and they solve exercises with few errors.
• 28–30 cum laude: the student has acquired all course content and is able to master and critically connect it; they solve exercises completely and without errors.

The individual project may add from 1 to 4 points to the base grade.

Assessment may also be conducted online if circumstances require.

Students with disabilities and/or specific learning disorders (DSA) should contact the instructor, the CInAP contact at DMI (Prof. Daniele), and CInAP well in advance to access appropriate compensatory measures

• Details on SIFT
• Self-calibration
• Intrinsic/Extrinsic Camera Parameters
• Viola/Jones Algorithm for Face Detection
• Image Generation Pipeline

It should be noted that these questions are purely indicative: the questions asked during the exam may differ, even significantly