Warning
This page is located in archive. Go to the latest version of this course pages. Go the latest version of this page.

Computer Vision Methods

Course Description

This course focuses on the following computer vision problems: finding correspondences between images using image features and their robust invariant descriptors, image retrieval, object detection and recognition, and visual tracking.

Pre-requisites

The course has no formal pre-requisits. However, certain skills and knowledge are assumed, and it is the responsibility of the student to get to the required level.

The assignments are implemented in the Python and numpy computing environment, and familiarity with it will help. The programing assignments, involving either implementing, modifying or testing computer vision methods, are a substantial part of the labs.

Knowledge of the basics of digital image processing as convolution, filtration, intensity transformations, image function interpolations and basic geometric transformations of the image (see the first lab) is assumed. Knowledge of linear algebra and probability theory is needed to understand the presented computer vision methods.

Lectures: Monday 11:00-12:30, KN:E-126

Lecturers: JM Jiří Matas, JC Jan Čech, DM Dmytro Mishkin, GT Giorgos Tolias, OD Ondřej Drbohlav,MS Milan Šulc

Lectures will be given via zoom, link: https://feectu.zoom.us/j/93818244152,
according the schedule, as listed below (from 11:00, zoom meeting opened at 10:50 )

Week Date Lecturer Slides Topic
1 15.2. JCDeep learning recording A shallow introduction into the deep machine learning. Convolutional Neural Networks. Principles, layers, architectures for image recognition.
2 22.2. JCDeep learning II recording* Deep architectures object detection and semantic segmentation. Further insights into the deep nets. Generative models (GANs).
3 1.3.JM, DM Correspondence 1st lecture slides*, recording Correspondences and wide baseline stereo. Motivation and applications. Interest point and distinguished regions detection: Harris operator (corner detection)
4 8.3.DM Correspondence 2nd lecture slides*, recording Laplace operator and its approximation by difference of Gaussians, Hessian detector, affine covariant version, Maximally Stable Extremal Regions (MSER).
5 15.3.DM Correspondence 3rd lecture slides*, recording Descriptors of measurement regions: SIFT (scale invariant feature transform), RootSIFT, shape context. LBP (local binary patterns), Matching. Deep learned features (HardNet).
6 22.3.DM, JM RANSAC-part1 recording Deep learned features (R2D2, SuperPoint): finish. Start of RANSAC.
7 29.3.JM, GT RANSAC-part2 recoding Retrieval-part1 recoding RANSAC: part 2. Image retrieval: task formulation, evaluation metrics, Bag-of-Words
8 5.4. Easter Monday
9 12.4.GT Retrieval-part2 recording Representation and matching models, spatial verification, query expansion, other tasks
10 19.4.GT Deep retrieval recording Transfer learning, architectures for global and local descriptors, pairwise loss, descriptor whitening
11 26.4.JM KLT, Mean Shiftrecording Tracking I. Introduction. Kanade-Lucas-Tomasi tracker. Mean Shift
12 3.5.JM KCF Tracking TLD, Tracking_by_Segmentationrecording Tracking II. KCF Kernel Correlation Filter. Long-term Tracking, TLD: Tracking-Learning-Detection, Tracking by Segmentation
13 10.5. JM Viola-Jones face detector, Waldboost,Hough Transform recording Object detection by sliding window and sequential decision making (Method of Viola and Jones, Waldboost. Hough transform and its comparison with RANSAC(Random Sample Consensus).
14 17.5. MS Recognition of Plants and Fungi recording Case study: Plant and Fungi recognition using Deep Nets

* Update of course slide material

* I forgot to start recording of the second lecture, I am sorry. The provided recording is from the last year. Note that the slides has changed slightly. Semantic segmentation architectures and deep fakes were added this year. Minor errors were fixed in the slides.

Evaluation

Work during the semester 50%, written part of the exam 40%, oral part of the exam 10%

Exam

Examples of exam questions. There will be 4-5 similar questions at the written part of the exam. The oral part of the questions takes place after the written part and will focused on discussion of your answers.

Literature

Lecture slides constitute the main source of study literature in this course.

Further Info

Further information is available in next sections of this page. We would appreciate your feedback on the contents and organization on the discussion forum of the course.



Good luck to all participants of the course.

Lecturers
jm_ct2008.11-3.jpg jancech.jpeg gtolias.jpeg dmytro.jpeg ondrej_drbohlav.jpg
Jiří Matas Jan Čech Giorgos Tolias Dmytro Mishkin Ondřej Drbohlav

Consultations are possible upon request.

courses/mpv/start.txt · Last modified: 2021/05/17 13:01 by sulcmila