Learning objectives
The course provides the necessary skills to design and develop computer vision systems, with a focus on state-of-art techniques based on the most commonly used software libraries and tools.
The course includes both theoretical notions and laboratory activities, where students will acquire competencies in software development, mainly in C/Linux/OpenCV environment.
Prerequisites
This course requires knowledge acquired in Calcolatori Elettronici, Fondamenti di Informatica, e Sistemi Operativi, as well as decent programming skills in C/C++
Course unit content
- - -
Full programme
Image acquisition, sampling, quantization
Image formats, video formats, acquisition
Preprocessing: pixel operation
Preprocessing: morphology
Lab on memory buffer manipulation
Preprocessing: local operation
Preprocessing: global operation
Lab on filtering and convolution
Edge Detection
Lab on Edge Detection
Camera 1
Camera 2
Camera 3
Lab on Single View Vision
Camera 4
Stereo Matching
Lab on Stereo Matching
Features and Descriptors 1
Features and Descriptors 2
Lab on Feature Matching and Model Fitting
Fitting and Matching
Lab on Feature Matching and Model Fitting
Introduction to Structure from Motion
Lab on SFM
Image Segmentation
Lab on Image Segmentation
Classification&Recognition 1
Classification&Recognition 2
Lab on Classification
Seminary of CNN
Bibliography
D. A. Forsyth and J. Ponce. Computer Vision: A Modern Approach (2nd Edition). Prentice Hall, 2011.
R. Hartley and A. Zisserman. Multiple View Geometry in Computer Vision. Cambridge University Press, 2003
Teaching methods
50% lectures, 50% laboratory
Assessment methods and criteria
Written exam on the theoretical part
Practical exam in Lab.
Alternatively to the final Practical exam, two mid-term partial exams will be offered to the student during the class period.
Other information
- - -