CS 6320 Computer Vision - Spring 2019

Instructor: Srikumar Ramalingam

Lecture Time: Mon,Wed 1:25PM-2:45PM

Place: WEB L114

TA: Xin Yu (xin.yu@utah.edu)

Abstract:

This class will provide the introduction to fundamental concepts in computer Vision. Topics in this class include
camera pose estimation, 3D reconstruction, feature detectors and descriptors, object recognition using vocabulary
tree, segmentation, stereo matching, graph cuts, belief propagation, and a brief introduction to deep neural networks.
In the assignments, the students will be expected to implement basic computer vision tasks such as segmentation,
stereo reconstruction, image matching using vocabulary tree, and small computer vision applications using deep
neural networks.

Class Schedule

Date Lecture Number Topic
1/7 L1 Introduction to Computer Vision
1/9 L2 Camera Models and Image Formation (Final)
1/14 L3 Camera Pose Estimation and RANSAC (Final)
1/16 L4 Class canceled & extra TA hours
1/21 Martin Luther King Jr. Day
1/23 L5 Image Matching (Final) ( Paper, Vocabulary Tree Example)
1/28 L6 Motion Estimation (Final)
1/30 L7 3D Reconstruction (Final)
2/4 L8 Introduction to Graphical Models (Final) Slides
2/6 L9 Introduction to Graphical Models (Final)
2/11 L10 Belief Propagation (Final)
2/13 L11 Stereo Matching
2/18 Presidents' Day
2/20 L12 Project Discussions
2/25 L13 Graph Cuts (Final)
2/27 L14 Graph Cuts (continued)
3/4 L15 Graph Cuts (continued)
3/6 L16 Object Detection
3/11 *Spring Break*
3/13 *Spring Break*
3/18 L17 Using neural nets to recognize handwritten digits (Final) ( Slides , Chapter1 )
3/20 L18 Using neural nets ... (continued)
3/25 L19 Project Discussions and Feedback
3/27 L20 How the backpropagation works (Final)
4/1 L21 How the backpropagation works ( Slides ,Chapter2)
4/3 L22 Improving the way the network learns ( Slides
4/8 L23 Convolutional Neural Networks ( Slides )
4/10 L24 Overview of Gradient Descent Algorithms ( Slides )
4/15 L25 Review (Pose Estimation, 3D Recon, Image Matching, Belief propagation) ( Practice Questions with answers )
4/17 L26 Review (Graph Cuts & Deep Learning) ( Practice Questions with Answers )
4/24 Final Exam (1:00 - 3:00 PM)
4/30 1 - 3 PM Final Project Presentations (5 + 2 minutes for each group)

Resources:

Programming Assignments

Assignments will be on topics assigned approximately every 3 weeks by the professor (i.e. there will be 4 assignments). You will be graded based on the best 4 out of 5 assignments. Assignments will require submission of the code in a zipped folder and findings in a pdf format, these will need to be submitted separately on the canvas. Programming is to be done in MATLAB (the basic package --- no extra toolkits).

Project

The project carries 20 points and it should clearly demonstrate your interest in computer vision. The problem statement carries 5 points. Implementation, results, and report will carry the remaining 15 points. The project could be an extension of one of the 5 programming assignments. For example, if you think of a cool new idea related to one of the assignments, you could implement the new idea and write a report on this idea. You can also choose a completely different project. You are free to do the project in a group, but the maximum allowed group size is 2. For a near-perfect project with great results and an excellent report, you can earn 5 points as extra credit.

Tests


Honor Policy

Students are expected to work on their own, as instructed by the Professor. Students may discuss projects with other individuals either in the class or outside the class, but they may not receive code or results electronically from any source that is not documented in their report. Any student who is found to be violating this policy will be given a failing grade for the course and will be reported to the authorities as described in the University's Student Code.

Accommodations Policy

The University of Utah seeks to provide equal access to its programs, services and activities for people with disabilities. If you will need accommodations in the class, reasonable prior notice needs to be given to the Center for Disability Services, 162 Olpin Union Building, 581-5020 (V/TDD). CDS will work with you and the instructor to make arrangements for accommodations. All written information in this course can be made available in alternative format with prior notification to the Center for Disability Services.

Grading