Computer Vision
Informacje ogólne
Kod przedmiotu: | 4010-CVI |
Kod Erasmus / ISCED: | (brak danych) / (brak danych) |
Nazwa przedmiotu: | Computer Vision |
Jednostka: | Interdyscyplinarne Centrum Modelowania Matematycznego i Komputerowego |
Grupy: | |
Punkty ECTS i inne: |
(brak)
|
Język prowadzenia: | angielski |
Rodzaj przedmiotu: | fakultatywne |
Skrócony opis: |
(tylko po angielsku) This course provides an overview of computer vision on modern Intel® architecture. Topics include: understanding how to use computer vision in industry; learning the main algorithms for image processing; exploring how machine learning is used in computer vision. |
Pełny opis: |
(tylko po angielsku) 1. History of computer vision. * How the modern industry uses computer vision * Significant technologies and libraries * Computer vision application development workflow 2. Core techniques of image processing. * Methods such as interpolation, color conversions, and thresholding * How to implement these tools in NumPy and Matplotlib 3. Image transformations and their usages. * How kernel methods (such as convolution) work and their impact on computer vision * How to use image gradients to perform edge detection * Learn more about edge detection by using Canny edge detectors * Understand more advanced techniques through Fourier transformations and wavelet methods 4. Image contours, segmentation, and image matching. * An overview of image contours and the techniques to find them * How segment the foreground and background to select areas for analysis 5. Image features and the techniques to find them. Learn about: * The methods to perform object detection and object recognition 6. Machine learning algorithms for computer vision. Topics include: * Support vector machines (SVM)—a popular algorithm used for classification problems * K-nearest neighbor clustering for image analysis * Unsupervised and supervised learning techniques 7. Application of neural networks. Learn about: * The mathematical theory supporting neural networks * How to use convolutional neural networks for image classification |
Literatura: |
Intel Academy: https://software.intel.com/en-us/ai/courses/computer-vision |
Efekty uczenia się: |
(tylko po angielsku) By the end of this course, students will have practical knowledge of: different techniques to process, transform, and classify images, how to apply deep learning to visual tasks, mportant computer vision methods, such as image segmentation and edge extraction. |
Metody i kryteria oceniania: |
Zaliczenie na podstawie samodzielnie opracowanego raportu na zadany temat oraz zaliczenia ćwiczeń (samodzielnej realizacji ćwiczeń w kursie on-line). |
Właścicielem praw autorskich jest Uniwersytet Warszawski.