TeachingWs09Cv2

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

3D Computer Vision II Winter Term 2009

Administrative Info

Lecture by Prof. Nassir Navab
Exercises by Diana Mateus and Loren Schwarz

2+2 SWS, 5 ECTS, Theoretische Informatik, Wahlpflichtfach

Time, Location & Requirements

Tuesday 10:30-12:00 MI 03.13.010
Exercises: Thursday 13:00-14:30 MI 03.13.010

Course information and certificate requirements:

  • The classes as well as the exercises will be held in English.
  • The lecture consists of one intermediate exam and a final exam.
  • All exams are hand written, the midterm is a closed-book exam (no notes allowed).

Office hours: Thursday 14:30-16:00 (MI 03.13.043)

Site Content

Announcements

  • The final attestation will take place on Tuesday, 09.02.2010, 10:30-12:00. You will be asked to present your homework results (from exercise sheet 5 on).
  • The final exam will take place on Friday, 26.02.2010, 14:00-15:30, in room MW 0250. Please register for the exam in TUMonline.

Overview

Making a computer see was something that leading experts in the field of Artificial Intelligence thought to be at the level of difficulty of a summer student's project back in the sixties. Forty years later the task is still unsolved and seems formidable. A whole field, called Computer Vision, has emerged as a discipline in itself with strong connections to mathematics and computer science and looser connections to physics, the psychology of perception and the neuro sciences.

Over the past decade there has been a rapid development in the understanding and modeling of the geometry of multiple views in computer vision. The theory and practice have now reached a level of maturity where excellent results can be achieved for problems that were unsolved a decade ago, and often thought unsolvable. These tasks and algorithms include problems like:
Given two/three/multiple images, and no further information, compute/estimate:

  • matches between the images
  • the 3D position of the points that generate these matches
  • the cameras that generate the images

adapted form Hartley & Zisserman's "Multiple View Geometry in Computer Vision"

This tasks and algorithms as well as the methods which allow us to reason about the quality of our results are the core of the lecture 3D Computer Vision II. This lecture deals with multiple view geometry problems and having attended 3D Computer Vision I is of great benefit, however, it is not, in any case, compulsory.

Content

  1. Intro, motivation & overview
  2. Projective geometry and 2D, 3D transformations
  3. Parameter estimation
  4. Camera model
  5. Two-View: Epipolar Geometry, 3D reconstruction, Fundamental Matrix, Structure computation
  6. Two-View: Mixed Camera Models: Perspective and Orthographic, Combined Othrographic and Perspective (COP) Images
  7. Three-View: The Trifocal Tensor
  8. Multi-View: Factorization Method, Multi-view Reconstruction, Motion and Structure from Motion

Midterm Exam Information

  • The exam will take place on Thursday, December 17 in the regular exercise hours (13:00-14:30) in our seminar room (MI 03.13.010).
  • The exam will not contain any MATLAB programming questions, as in the homeworks.
  • There will be more theoretical questions on basic concepts that you have learned so far in the lecture.
  • The exam is a closed-book exam: you are not allowed to have any notes with you (no hand-written papers either).
  • Don't worry, you are not asked to memorize complicated formulae. If you understand the main concepts and their defining equations, you should be fine.
  • There will be multiple-choice questions and small problems to solve, similar to those we did in the exercise sessions.

Course Schedule

Please scroll down for the exercise schedule.

Date Topic Conducted by Material/Literature
Thu, Oct 22 Introduction Prof. Dr. N. Navab Introduction
Tue, Oct 27 Reminder: Projective Geometry, Transformations, Camera Models Prof. Dr. N. Navab Transformations
Camera Models
Tue, Nov 3 Two-View Geometry (Epipolar Geometry) Prof. Dr. N. Navab Epipolar Geometry
Thu, Nov 12 Two-View Geometry (Estimation of Fundamental Matrix) Prof. Dr. N. Navab Fundamental Matrix
Tue, Nov 17 no lecture due to Studentische Vollversammlung    
Tue, Nov 24 Two-View Geometry (continued) Prof. Dr. N. Navab  
Thu, Nov 26 Stereo I (Rectification and Disparity) Christian Unger Rectification
Tue, Dec 8 Stereo II (Matching, Reconstruction) Prof. Dr. N. Navab Stereo Matching
Tue, Dec 15 Combined Orthographic and Perspective Camera Models Prof. Dr. N. Navab COP Imaging
Tue, Dec 22 Photometric and Geometric Information for Pose Estimation Pierre Georgel Lecture Slides
Tue, Jan 12 Multi-View 3D-Reconstruction Alexander Ladikos Multi-View 3D-Reconstruction
Tue, Jan 19 Factorization Diana Mateus Factorization
Tue, Jan 26 Three-view Geometry and Tri-focal Tensor Prof. Dr. N. Navab Trifocal Tensor
Tue, Feb 2 Simultaneous Localization and Mapping (SLAM) Stefan Holzer SLAM

Exercise Schedule

Date Topic Conducted by Exercises Solutions Homework due
Tue, Oct 20 Introduction, Registration D. Mateus, L. Schwarz      
Thu, Oct 29 Projective Geometry, Camera Models D. Mateus, L. Schwarz     Midterm Attestation
Thu, Nov 5 Epipolar Geometry D. Mateus, L. Schwarz     Midterm Attestation
Tue, Nov 10 Estimation of Fundamental Matrix D. Mateus, L. Schwarz     Midterm Attestation
Thu, Nov 19 Rectification D. Mateus, L. Schwarz     Midterm Attestation
Tue, Dec 1 Rectification D. Mateus, L. Schwarz     Midterm Attestation
Thu, Dec 3 Stereo Matching D. Mateus, L. Schwarz     Final Attestation
Thu, Dec 10 Attestation D. Mateus, L. Schwarz      
Thu, Dec 17 Midterm Exam        
Thu, Jan 7 Midterm Exam Correction D. Mateus, L. Schwarz      
Thu, Jan 14 Multi-View Reconstruction D. Mateus, L. Schwarz     Final Attestation
Thu, Jan 21 Factorization D. Mateus, L. Schwarz     Final Attestation
Thu, Jan 28 Trifocal Tensor D. Mateus, L. Schwarz     Final Attestation
Thu, Feb 4 SLAM Stefan Holzer      
Tue, Feb 9 Final Attestation        


TeachingForm
Title: 3D Computer Vision II
Professor: Prof. Dr. Nassir Navab
Tutors: Diana Mateus and Loren Schwarz
Type: Lecture
Information: 2+2 SWS, 5 ECTS, Theoretische Informatik, Wahlpflichtfach
Term: 2009WiSe
Abstract: Making a computer see was something that leading experts in the field of Artificial Intelligence thought to be at the level of difficulty of a summer student's project back in the sixties. Forty years later the task is still unsolved and seems formidable. A whole field, called Computer Vision, has emerged as a discipline in itself with strong connections to mathematics and computer science and looser connections to physics, the psychology of perception and the neuro sciences. Over the past decade there has been a rapid development in the understanding and modelling of the geometry of multiple views in computer vision. The theory and practice have now reached a level of maturity where excellent results can be achieved for problems that were unsolved a decade ago, and often thought unsolvable. These tasks and algorithms include problems like: Given two/three/multiple images, and no further information, compute/estimate: * matches between the images * the 3D position of the points that generate these matches * the cameras that generate the images adapted form Hartley & Zisserman's "Multiple View Geometry in Computer Vision" This tasks and algorithms as well as the methods which allow us to reason about the quality of our results are the core of the lecture 3D Computer Vision II. This lecture deals with multiple view geometry problems and having attended 3D Computer Vision I is of great benefit, however, it is not, in any case, compulsory.


Edit | Attach | Refresh | Diffs | More | Revision r1.75 - 10 Feb 2011 - 13:55 - LorenSchwarz

Lehrstuhl für Computer Aided Medical Procedures & Augmented Reality    rss.gif