This course covers digital image processing as well as advanced topics in computer vision. Initial topics include image formation, digital filtering, sensor modelling and feature detection techniques. The course will discuss how these algorithms are used to address general computer vision problems including three-dimensional reconstruction, scene understanding, object recognition, and motion analysis.
CS 536 or consent of instructor.
Students should have a solid background in linear algebra, a familiarity with numerical techniques, and geometry. Basic programming skills are required.
Students will learn both the fundamental issues and advanced techniques related to computer vision and knowledge-directed processing of digital images. The primary concepts learned will be the fundamentals of digital image formation, camera calibration, multi-camera system issues such as epipolar geometry, dynamic view warping, motion analysis, and feature extraction and analysis. Underlying these skills, students will be introduced to concepts such as surface representation, active contours, geometric hashing, and model fitting. Finally, students will understand how these techniques and concepts fit together in successful computer vision systems that support "intelligent" tasks, such as path planning, navigation and road following, object recognition, learning, and scene classification.
Introduction to Computer Vision
Image formation, Image geometry
Human vs. Machine Perception
Digital Image Processing
Convolution and Noise Removal
Computer Vision Topics
Pose determination, registration
Motion Analysis: Structure from motion
Surface Modeling, Range Sensors
Regression Analysis (Surface Fitting, refinement)
Knowledge-Based Computer Vision
Knowledge Representation (example systems)
Uncertainty and Belief
Learning and Vision
Gaze Control, Visual Search
Cooperative Vision (multi-agent)
Exact details about examinations in this course will be determined by the instructor offering the course. Typically there will be two in-class examinations during the semester and a two-hour final examination. Specific details will be made available in the syllabus at the start of each semester in which the course is offered.
A student's grade will be determined by a weighted average of homework assignments, programming exercises, projects, midterm examinations, and the final examination. The faculty offering the course will make the details available at the start of the course. A typical weighting is:
Homeworks and programs: 40%
Midterm Examinations (2 @ 15%): 30%
Final Examination: 30%
Introductory Techniques for Three-Dimensional Computer Vision,
E. Trucco and Verri,
Prentice-Hall, 1998. ISBN: 0-13-261108-2
High-Level Vision: Object Recognition and Visual Cognition,
MIT Press, 1992. ISBN: 0262210134.