Multimedia Information Processing Group

Inf-CV: Image-based 3D Scene Reconstruction


The students learn to handle entities of projective geometry and image-based geometric transformations and implement these in the context of image-based 3-D scene reconstruction. Programming exercises are solved with the help of OCTAVE. The following topics are discussed:

  • Image sequence correspondence analysis
  • Basics of projective geometry
  • Homographies and panoramic images from rotating cameras
  • Multi-view geometry from a moving camera
  • Epipolar geometry and depth estimation
  • Camera tracking and pose estimation
  • Application in the field of augmented reality and image-based modeling
All lecture slides and course material will be in English. The lecture will be held in English.




Basic mathematical knowledge from Bachelor courses in linear algebra, 3-D geometry and solving of linear equations. Basic knowledge in image processing, like the Bachelor lecture InfEinfBV (Introduction to Image Processing) is requested. IMPORTANT: Due to the Corona Crisis, the course will start online and all materials like course videos, lecture notes and excercises will be handled by OLAT. We will provide online lecture videos, online excercise lectures and feedback sessions, and homework. You are requested to register at both OLAT and StudiDB until April 6th. 1. Please register for this course at StudiDB 2. Please register for course materials at the OLAT 3D Scene Reconstruction course

Recommended literature

Literature: Szeliski, Rick: Computer Vision: Algorithms and Applications. Springer 12010. Elektronische Version: Hartley, Zissermann: Multiple View Geometry, Cambridge 2003. Schreer: Stereoanalyse und Bildsynthese, Springer 2004

Additional information