ENGINEERING TRIPOS PART IIB – 2012/2013

Module 4F12 - Computer Vision and Robotics

AIMS

The module aims to introduce the principles, models and applications of computer vision. The course will cover image structure, projection, stereo vision, and the interpretation of visual motion. It will be illustrated with case studies of industrial (robotic) applications of computer vision, including visual navigation for autonomous robots, robot hand-eye coordination and novel man-machine interfaces.

LECTURE SYLLABUS

• Introduction (1L)
  Computer vision: what is it, why study it and how ? The eye and the camera, vision as an information processing task. A geometrical framework for vision. 3D interpretation of 2D images. Applications.
  
  
• Image structure (2L)
  Image intensities and structure: edges and corners. Edge detection, the aperture problem. Corner detection. Contour extraction using B-spline snakes. Case study: tracking edges and corners for robot hand-eye coordination and man-machine interfaces.
  
  
• Projection (4L)
  Orthographic projection. Pin-hole camera model. Planar perspective projection. Vanishing points and lines. Projection matrix, homogeneous coordinates. Camera calibration, recovery of world position. Weak perspective, the affine camera. Projective invariants. Case study: 3D models from uncalibrated images using PhotoBuilder.
  
  
• Stereo vision (2L)
  Epipolar geometry and the essential matrix. Recovery of depth. Uncalibrated cameras and the fundamental matrix. The correspondence problem. Affine stereo. Case study: 3D stereograms.
  
  
• Object detection and tracking (4L, Prof A. Blake and Prof R. Cipolla)
  Basic target tracking; Kalman filter; application to B-spline snake. Active appearance models. Chamfer matching, template trees. Case study: intelligent automotive vision system.
  
  
• Example classes (3L, Prof R. Cipolla)
  Discussion of examples papers and past examination papers.

OBJECTIVES

On completion of the module, students should:

• Be able to design feature detectors to detect, localise and track image features;
• Know how to model perspective image formation and calibrate single and multiple camera systems;
• Be able to recover 3D position and shape information from arbitrary viewpoints;
• Appreciate the problems in finding corresponding features in different viewpoints;
• Analyse visual motion to recover scene structure and viewer motion, and understand how this information can be used for navigation;
• Understand how simple object recognition systems can be designed so that they are independent of lighting and camera viewpoint;
• Appreciate the industrial potential of computer vision but understand the limitations of current methods.

REFERENCES

Please see the Booklist for Group F Courses for references for this module.

Last updated: June 2012