Stereo Vision, Dense Motion & Tracking

This course introduces learners to stereo vision theory, dense motion and visual tracking. Topics include epipolar geometry, optical flow, structure from motion, multi-object tracking, 3D vision and visual odometry. Learners gain experience writing computer vision programs through online labs using MATLAB and supporting toolboxes.

This is the third course in the Computer Vision series that lays the groundwork necessary for designing sophisticated vision applications.

This online course is available on the Coursera platform.

Course Duration: Approximately 4 weeks (4-5 hours of effort per week)

Learning Outcomes

  • Develop 3D vision applications using a stereo imaging system
  • Understand epipolar geometry
  • Understand motion perception, motion field and optical flow
  • Discuss the techniques used to obtain the 3D structure of objects

Intended Audience

Anyone curious about or interested in exploring the concepts of stereo and 3D vision, dense motion and visual tracking


  • Basic programming skills & experience, specifically in MATLAB
  • Familiarity with basic linear algebra, calculus & probability, and 3D co-ordinate systems & transformations