Artificial Vision
A.Y. 2022/2023
Learning objectives
Aim of the course is to give elements which allow to infer knowledge about the real world from digital images or videos. These skills mainly concern the 3D reconstruction of real objects, and the processing and recognition of elements and actions in a scene.
Expected learning outcomes
- Learn the image formation principles
- Learn the techniques to obtai 3D object reconstruction
- Learn machine learning techniques aimed at the identification and recognition of objects and actions in images or videos
- Learn the techniques to obtai 3D object reconstruction
- Learn machine learning techniques aimed at the identification and recognition of objects and actions in images or videos
Lesson period: Second semester
Assessment methods: Esame
Assessment result: voto verbalizzato in trentesimi
Single course
This course cannot be attended as a single course. Please check our list of single courses to find the ones available for enrolment.
Course syllabus and organization
Single session
Responsible
Lesson period
Second semester
Lectures will be held both in presence and in video streaming through the Zoom platform.
Material:
the program and the material will not change with respect to the standard one.
Material:
the program and the material will not change with respect to the standard one.
Course syllabus
Aim of this course is to examine the fundamental concepts in the field of computer vision, with special focus on the following topics:
* Image formation
- Geometric camera models
- Camera Calibration
- Brightness models
- Color models
* "Early" vision
- Linear Filtering
- Feature extraction
- Stereopsis (binocular vision)
- Structure from Motion (multi-view)
- 3D Registration
* Machine learning
- Linear Neural Networks
- Multilayer perceptrons
- Deep Learning Computation
- Convolutional Neural Networks
- Recurrent Neural networks
- Generative models
* Image formation
- Geometric camera models
- Camera Calibration
- Brightness models
- Color models
* "Early" vision
- Linear Filtering
- Feature extraction
- Stereopsis (binocular vision)
- Structure from Motion (multi-view)
- 3D Registration
* Machine learning
- Linear Neural Networks
- Multilayer perceptrons
- Deep Learning Computation
- Convolutional Neural Networks
- Recurrent Neural networks
- Generative models
Prerequisites for admission
A good knowledge of the fundamentals of:
- probability and statistics
- signal and image processing
as taught in scientific undergraduate courses.
- probability and statistics
- signal and image processing
as taught in scientific undergraduate courses.
Teaching methods
Lectures and class exercises.
Teaching Resources
Web site:
https://github.com/lanzarotti/
Class material:
- Lecture slides
- Software code for exercises
Testo di riferimento:
- D.A. Forsyth, J. Ponce - Computer Vision - A Modern Approach - Pearson, 2nd edition
- "Dive into Deep Learning" (https://d2l.ai)
https://github.com/lanzarotti/
Class material:
- Lecture slides
- Software code for exercises
Testo di riferimento:
- D.A. Forsyth, J. Ponce - Computer Vision - A Modern Approach - Pearson, 2nd edition
- "Dive into Deep Learning" (https://d2l.ai)
Assessment methods and Criteria
The exam consists of an oral evaluation.
Optionally students can carry out a project to deepen the knowledge.
Optionally students can carry out a project to deepen the knowledge.
Educational website(s)
Professor(s)