|
CS484/555 - Introduction to Computer Vision - Spring 2024
ANNOUNCEMENTS
This is the website for CS484/555, Spring Semester 2024.
Course materials, slides, assignments, etc. will be accessible via Moodle.
Make sure you visit this page often enough as any further announcements will be made here.
GenAI Policy
Using AI models for any coursework including assignments, projects, exams, and quizzes is completely prohibited unless specifically designated and approved by the lecturer.
CLASS HOURS
Wednesday: 09:30-10:20, Friday: 13:30-15:20 in B-203.
LECTURER
Shervin R. Arashloo
Office: EA-429
Office Hour: By appointment
E-mail: s.rahimzadeh-at-cs.bilkent.edu.tr
Teaching Assistants
TBA
COURSE DESCRIPTION
Image acquisition, sampling and quantization. Spatial domain processing. Image enhancement. Texture analysis. Edge detection. Frequency domain processing. Color image processing. Mathematical morphology. Image segmentation and region representations. Statistical and structural scene descriptions. Applications. Credits: 3 units
COURSE EMPHASIS AND GOALS
This course provides an introduction to image analysis and computer vision. We will start with low-level vision (early processing) techniques such as binary image analysis, filtering, edge detection and texture analysis. Then, we will cover mid-level vision topics such as image segmentation and feature extraction in detail. Finally, we will do case studies on several applications such as image classification, object recognition, and deep learning.
PREREQUISITES
(CS 102 or CS 114 or CS 115) and (MATH 225 or MATH 220 or MATH 241) and (MATH 230 or MATH 255 or MATH 260)
Students taking this course need to have good background on high-level programming, data structures, linear algebra, and matrix calculus.
No prior knowledge of image processing or computer vision is assumed.
TEXTBOOKS
R. Szeliski, Computer Vision: Algorithms and Applications, Springer 2010. (local copy)
R. C. Gonzalez and R. E. Woods, Digital Image Processing, 4th edition, Pearson, 2018.
D. A. Forsyth and J. Ponce, Computer Vision: A Modern Approach, 2nd edition, Pearson, 2012.
D. H. Ballard and C. M. Brown, Computer Vision, Prentice Hall, 1982.
L. G. Shapiro and G. C. Stockman, Computer Vision, Prentice Hall, 2001.
I. Goodfellow, Y. Bengio and A. Courville, Deep Learning, MIT Press, 2016.
EXAMS
The midterm exam will be held at ? during ? on ?. The exam will cover all topics from the beginning of the semester until the end of ? chapter. You are allowed to bring only the lecture notes (slides) without any additional notes.
There will be several quizzes with/without prior notice throughout the semester. There will be no make-ups for the missed quizzes.
HOMEWORK
There will be three homework assignments. These will be posted on Moodle.
PROJECT
The goal of the project is to develop/implement a computer vision algorithm and assess its performance on relevant and standard data.
You will work in a group of two. (If there is an odd number of students, one group will consist of three students.)
Although there is no restriction for the topic that you will select (as long as it is related with the course contents), the students should seek consent for their project topic. Since we want to minimise possible overlaps over project topics, two groups will not be allowed to work on a very similar topic; The project topics will be confirmed on a first-come-first-served basis.
You will have three options for the term project:
Choose one paper relevant to the course contents and then implement the algorithm proposed in that paper. Do not use any codes provided by the authors of the paper, or any other people, in case available. Run the algorithm you implemented on the data set you will select and compare and report the results using standard performance metrics, also using statistical tests. Additionally, follow a proper way of selecting the algorithms’ (hyper)parameters and also conduct parameter analysis. I expect you to select a recent paper that explains a not-so-straightforward algorithm.
Train and evaluate a deep learning model for a data set you will select. Here you may use the third-party codes, but you CANNOT select any data sets that was used to pre-train any of the deep learning models (e.g., you cannot use the ImageNet data set to conduct your experiments). In this option, you are expected to get the model trained on your data set and obtain reasonable test set performances. An important part of this second option, is to explore the effects of different (hyper)parameters on the performance of the model. You need to select a deep neural network and report its performance on your data set. You are expected to select a not-so-easy data set.
Presentation and Reporting
You must prepare and submit a final report for your project along with the developed codes/trained models as two separate files (a pdf file for the report and a single archive file (e.g., zip, tar, rar) for the code) by ? on ?. The reports are expected to be around 6 pages and must follow the IEEE two-column format as described in their templates. Try to follow the format as closely as possible. Both the content and the format will be subject to grading.
Each group will give a presentation on their project in class. Every group member should take a part in presentation. You will have approximately 10-12 minutes for your presentation; we will have a discussion period of 5 minutes after the presentation. I will let you know the exact duration after the add-drop period.
The presentation content, its format and layout, and the way that you present it will affect your grade. The interest that your presentation attracts from the audience will also affect your grade.
Prepare your slides neatly and properly. It should contain at most 12 slides with reasonable content (only present the most important and interesting parts). Do not copy and paste any text/equation/table from a paper (if necessary, type them). If you need to use a figure (or an image) of a paper, take it but give a credit to this paper (so that we can understand how much effort you have put in preparing your presentation).
Deadlines for the project presentation and report
You will lose points if you miss the deadlines.
Feb. 14: Your group preference, if any
Feb 28: Term project selection (as a group)
March 29: Progress report (tentative)
April 21: Project presentations (tentative)
May 17: Final report for the project
Late submissions
Any assignments turned in late will be penalised and will incur a reduction of 25% in the final score, for each day (or part thereof) it is late. We follow no extension policy.
Please make sure you fully understand the code of discipline at Bilkent University and the Rules and Regulations of the Higher Education Council (YOK) (in Turkish). Cheating and plagiarism on exams, quizzes, and assignments will be punished according to these regulations.
GRADING POLICY
|