3D Vision (EE 569)

2014 Fall
Faculty of Engineering and Natural Sciences
Electronics Engineering(EE)
3
10
Mustafa Ünel munel@sabanciuniv.edu,
Click here to view.
English
Doctoral, Master
--
Formal lecture
Click here to view.

CONTENT

This course deals with a central problem in vision -how to recover 3-D structure and motion from a collection of 2-D images-using techniques mainly from linear algebra and matrix theory. Topics included are: geometric image formation, camera models, image feature extraction and tracking, camera calibration, stereo, epipolar geometry, eight-point algorithm, 3D reconstruction from two or more images, motion estimation etc. The aim is to provide graduate students in EE, CS and ME with a solid theoretical and algorithmic background for research.

OBJECTIVE

Objective of the course is to develop a unified framework for studying the geometry of multiple images of a 3-D scene and reconstructing geometric models from those images. Additionally the course provides graduate students with a solid theoretical and algorithmic background for research.

LEARNING OUTCOMES

  • On successful completion of the course, students will be able to: - Use various camera models in different applications
  • - Extract important image features and track them in real-time
  • - Compute optical flow
  • - Calibrate cameras
  • - Estimate camera motion and 3D structure
  • - Understand basics of multi-view geometry

ASSESSMENT METHODS and CRITERIA

  Percentage (%)
Individual Project 50
Homework 50

RECOMENDED or REQUIRED READINGS

Textbook

An Invitation to 3-D Vision: From Images to Geometric Models, Y. Ma, S. Soatto, J. Kosecka, S. S. Sastry, Springer, 2005

Optional Readings

- Computer Vision: Algorithms and Applications, R. Szeliski, Springer, 2010
- Multiple View Geometry in Computer Vision, 2nd Edition, R. Hartley and A. Zisserman, Cambridge Univ. Press, 2003
- Computer Vision: A Modern Approach, D. A. Forsyth and J. Ponce, Prentice-Hall, 2003 (available as pdf)