This course provides an introduction to computer vision, an area that lies in the heart of many modern AI applications: robotics navigation, searching through billions of images, autonomous vehicles and social networking. Computer vision technologies can understand and reconstruct the visual world. The course covers a range of topics, including feature detection, motion estimation, panoramas, 3D shape reconstruction, and object detection and recognition. This course emphasizes hands-on experience with computer vision, with several large programming projects.
The course will be self-contained. Students do not need to have computer vision background, however, knowledge of the following is required:
  • Working knowledge of MATLAB or Python and preferably also C/C++
  • Linear algebra
  • Basic probability theory and statistics
  • Recommended: Vector calculus, Data structures
The course will have readings from Computer Vision: Algorithms and Applications (available online), by Richard Szeliski.