AIM 5014: Computer Vision

Graduate course, Yeshiva University, MS in Artificial Intelligence, 2022

2022 Fall


Image analysis and computer vision are important to enable machines (computers) to understand the world and have become extremely important and popular in recent years. This course focuses on fundamental concepts, advanced topics, and interests in image analysis and computer vision. Topics to be covered include image formation, image filtering, edge detection, image classification/segmentation/regression, convolution neural network, deep learning, objection detection, semantic segmentation, etc. This course will enable students to apply computer vision and image processing techniques to solve various medical and industrial problems and learn to apply these techniques in research or industry.


By the end of this course, students will be able to:

● Understand the major concepts in image analysis and computer vision

● Know a range of computer vision techniques and where to apply them

● Understand state-of-the-art of deep learning architectures

● Obtain practical experience in implementing real-world projects and build their own models for effective problem solving



● Projects are to be completed and graded in Python. Basic programming skills in Python are required.

Required Texts:

● Richard, Szeliski. Computer Vision: Algorithms and Applications, 2nd edition (2021).

Additional readings (either web-based or provided by the instructor) will be assigned.


Discussions / Response Assignments: Discussions will focus primarily on use cases, papers, podcasts, and presentations related to computer vision.10%
Projects (3): Students will work individually and in teams on code-based “mini-projects” related to the current material. At the end of the course, each student will have a portfolio of relevant projects ready to show an employer.45%
Final Project Proposal, Project, Paperwork, and Presentation: Working individually or as part of a small team, students will build an artificial intelligence project that compares approaches to address a challenging and worthwhile problem of their choosing. Project evaluation criteria will include evaluation of (1) the statement of a compelling hypothesis or guiding question, (2) relevant choice of data sets(s), (3) comparative evaluation of different models, parameters, and hyper-parameters (4) findings and recommendations, and (5) a code notebook that is well-organized, clearly written, consistent with best practices, and reproducible. Students will present their final projects to their peers for feedback before submitting their final project write-up. The final presentation should be delivered within specified time parameters and include the appropriate level of detail for its intended audience. The final paperwork is targeted for publication purposes.45%


Quality of PerformanceLetter GradeRange %GPA/ Quality Pts.
Excellent - work is of exceptional qualityA93 - 1004
A-90 - 92.93.7
Good - work is above averageB+87 - 89.93.3
SatisfactoryB83 - 86.93
Below AverageB-80 - 82.92.7
PoorC+77 - 79.92.3
C70 - 76.9 2
FailureF< 700