COMPUTER VISION
Course objectives
General goals: The course aims at introducing students to a wide-spectrum presentation of Computer Vision. Specific goals: The course aims at providing the basic principles, methodologies and algorithms used for the design and application of computer vision systems Knowledge and understanding: Introductions of the fundamental principles and different areas of Computer Vision and knowledge on problem solving such as feature extraction, tracking, scene analysis, object recognition, event analysis, emotion analysis. Applying knowledge and understanding: The successful student will be able to exploit the portfolio of techniques and the different approaches shown in the course for the design and the successful implementation of vision systems. Critical and judgmental abilities Students will learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods in the design of vision system. Communication skills: Students will be able to interact proficiently with other Computer Vision researchers on a wide set of AI topics. Learning abilities: Students will be able to extend their skills in the subjects of this course, by the autonomous reading of the scientific literature on Computer Vision.
Channel 1
MARCO RAOUL MARINI
Lecturers' profile
Program - Frequency - Exams
Course program
The course explores basic image processing techniques, from the physical foundations of light to machine/deep learning to extract features for recognizing actions performed by subjects in a scene. The curriculum follows the logical thread outlined in the course textbook and is subject to change each year, especially in the final section, to demonstrate new and more competitive techniques for solving specific real-world problems.
Prerequisites
Only few prerequisites are needed:
Programming experience in C/C++, Python (mostly), or similar languages.
Mathematical foundations in linear algebra, calculus, and statistics.
Books
Primary Textbook:
Computer Vision: A Modern Approach, 2nd Edition – David Forsyth and Jean Ponce.
Optional Textbooks:
Computer Vision: Algorithms and Applications – Richard Szeliski.
*Introductory Techniques for 3-D Computer Vision* – E. Trucco and A. Verri.
Digital Image Processing – R.C. Gonzalez and R.E. Woods.
Additional Resources:
Vision Science: Photons to Phenomenology – Stephen E. Palmer (for color and perception).
Research papers from conferences like CVPR, ICCV, and journals like IEEE TPAMI.
Frequency
Attendance is not mandatory but strongly recommended.
Exam mode
The examination consists of two components: a slide-based presentation of a selected topic and a demonstration of a project aligned with the paper’s topic, with an associated report. The maximum duration for each component is 10 minutes (10 minutes for the paper and 10 minutes for the project). Group work is permitted, with a maximum of 4 students per group. Topics may be assigned by proposing a specific topic or problem within a chosen field of interest (direct request) or by requesting a suggested topic if no specific idea is identified (indirect request). The project requires basic coding skills, and pre-existing code may be used, but it must be adapted to align with the paper’s topic and the problem being addressed. Original system development is not mandatory. Code and presentation materials must be submitted two days before the examination date. The final grade is determined by course attendance (5%), the final presentation (40%), and the final project (55%). Additional considerations, such as creativity, submission of a supplementary paper, and task complexity, may contribute up to approximately 10%. Final grades reflect the effort and rigor demonstrated throughout the examination.
Bibliography
Computer Vision: A Modern Approach, 2nd Edition – David Forsyth and Jean Ponce.
Presented slides during the course.
Lesson mode
Classes are held in person and often require the use of your own laptop.
MARCO RAOUL MARINI
Lecturers' profile
Program - Frequency - Exams
Course program
The course explores basic image processing techniques, from the physical foundations of light to machine/deep learning to extract features for recognizing actions performed by subjects in a scene. The curriculum follows the logical thread outlined in the course textbook and is subject to change each year, especially in the final section, to demonstrate new and more competitive techniques for solving specific real-world problems.
Prerequisites
Only few prerequisites are needed:
Programming experience in C/C++, Python (mostly), or similar languages.
Mathematical foundations in linear algebra, calculus, and statistics.
Books
Primary Textbook:
Computer Vision: A Modern Approach, 2nd Edition – David Forsyth and Jean Ponce.
Optional Textbooks:
Computer Vision: Algorithms and Applications – Richard Szeliski.
*Introductory Techniques for 3-D Computer Vision* – E. Trucco and A. Verri.
Digital Image Processing – R.C. Gonzalez and R.E. Woods.
Additional Resources:
Vision Science: Photons to Phenomenology – Stephen E. Palmer (for color and perception).
Research papers from conferences like CVPR, ICCV, and journals like IEEE TPAMI.
Frequency
Attendance is not mandatory but strongly recommended.
Exam mode
The examination consists of two components: a slide-based presentation of a selected topic and a demonstration of a project aligned with the paper’s topic, with an associated report. The maximum duration for each component is 10 minutes (10 minutes for the paper and 10 minutes for the project). Group work is permitted, with a maximum of 4 students per group. Topics may be assigned by proposing a specific topic or problem within a chosen field of interest (direct request) or by requesting a suggested topic if no specific idea is identified (indirect request). The project requires basic coding skills, and pre-existing code may be used, but it must be adapted to align with the paper’s topic and the problem being addressed. Original system development is not mandatory. Code and presentation materials must be submitted two days before the examination date. The final grade is determined by course attendance (5%), the final presentation (40%), and the final project (55%). Additional considerations, such as creativity, submission of a supplementary paper, and task complexity, may contribute up to approximately 10%. Final grades reflect the effort and rigor demonstrated throughout the examination.
Bibliography
Computer Vision: A Modern Approach, 2nd Edition – David Forsyth and Jean Ponce.
Presented slides during the course.
Lesson mode
Classes are held in person and often require the use of your own laptop.
- Lesson code1047618
- Academic year2025/2026
- CourseComputer Science
- CurriculumSingle curriculum
- Year2nd year
- Semester1st semester
- SSDINF/01
- CFU6