Instructor: Simen Haugo (simen.haugo@ntnu.no)

Course information The course explores the state-of-the-art of various computer vision topics via guided self-study through a selection of papers. Each topic and its related papers gets introduced during the first half of the semester with its own short lecture. The students form groups and make a presentation about one or more papers from one topic of their choice. The topic is then revisited later in the semester on a day dedicated to the topic, with student presentations and discussions. The final topic list may change, but the following is a preliminary topic list for 2022:

• Camera modeling and calibration
• Visual localization and place recognition
• 3D model fitting/registration and tracking
• Feature extraction, matching and tracking
• Triangulation, relative pose and absolute pose estimation

Tasks during the course: The students are expected to work with questions and themes connected to their chosen papers and create a presentation.

Teaching form: Guided self-study based on selected literature and introductory lectures. The students form study groups and may request further guidance in a scheduled meeting if desired.

Schedule: The topics are introduced on a weekly basis during the beginning of the semester, with the corresponding student presentation day being held 6-7 weeks after the topic was introduced. The first topic and paper selection will be introduced in Week 35.

Evaluation: Individual oral exams. The basis for the exam is the student's own presentation, and in part the presentations of other students.

Information: Time: TBA (week 35) Lecture 1: TBA (week 35)