Course Identification

Title:

Multiple View Geometry for Computer Vision Applications

Code:

20214042

Lecturers and Teaching Assistants

Lecturers:

Prof. Ronen Basri, Dr. Meirav Galun

TA's:

Amnon Geifman, Hodaya Koslowsky, Dr. Yoni Kasten

Course Schedule and Location

Year:

2021

Semester:

Second Semester

When / Where:

Tuesday, 09:15 - 11:00

First Lecture:

23/03/2021

End date:

31/08/2021

Field of Study, Course Type and Credit Points

Mathematics and Computer Science: Lecture; Elective; Regular; 2.00 points

Comments

N/A

Prerequisites

Introduction to computer vision (given in the first semester)

Elements of linear algebra (e.g., Basic topics, by H. Dym)

Basic acquaintance with Matlab

Restrictions

Participants:

50

Language of Instruction

English

Attendance and participation

Required in at least 80% of the lectures

Grade Type

Numerical (out of 100)

Grade Breakdown (in %)

Assignments:

40%

Final:

60%

Evaluation Type

Final assignment

Scheduled date 1

Date / due date

N/A

Location

N/A

Time

-

Remarks

N/A

Estimated Weekly Independent Workload (in hours)

4

Syllabus

The course will focus on understanding the geometry of cameras in 3D, multi-view settings, and the related optimization and deep-learning-based algorithms. The following topics will be covered: epipolar geometry, fundamental and essential matrices, triangulation, camera calibration, depth estimation, bundle adjustment, robust estimation and others. The course involves programming and theoretical assignments and a final project.

Learning Outcomes

The students will gain "hands-on" practice in the area of 3D computer vision and structure from motion, including theory and common tools, along with modern techniques based on deep-learning.

Reading List

Multiple view geometry in computer vision, by Hartley and Zisserman

Website

N/A