The lecture introduces the basic concepts of image formation, perspective projection, and camera motion. The goal is to reconstruct the three-dimensional world and camera motion from multiple images. To this end, we study correspondences between points in different images and the geometric constraints that enable the recovery of motion and 3D structure. Particular emphasis is placed on the mathematical description of rigid body motion and perspective projection. Methods for estimating camera motion and 3D geometry include both spectral techniques and nonlinear optimization, complemented by practical aspects of 3D vision as well as recent advances and deep learning approaches in the field.

If you plan to attend, please register for the course on TUMonline. Later during the semester, you will have to register for the exam.

You can discuss course contents and receive announcements via Moodle.

For personal inquiries, please get in touch with us via 3dcv-ss26@vision.in.tum.de. This email reaches the entire course team. Please refrain from contacting team members individually.

Time and Location:

• Wednesdays: 102, Hörsaal 2, "Interims I" (5620.01.102), 12:00 - 14:00
  
• Thursdays: 00.02.001, MI HS 1, Friedrich L. Bauer Hörsaal (5602.EG.001), 11:00 - 12:00

Lecturer: Prof. Dr. Daniel Cremers
Start: Wednesday, 21.04.2025
End: Thursday, XX.07.2025

Time and Location:

• Wednesdays: 102, Hörsaal 2, "Interims I" (5620.01.102), 16:00 - 18:15

Start: Wednesday, 28.04.2025

Teaching Assistants:
Daniil Sinitsyn

This year, all the lectures and exercise sessions will be recorded and streamed automatically via TUM Live: https://live.rbg.tum.de/