Computer Vision is the study of inferring properties of the 3D world based on one or more 2D digital images or video sequences.

This is an advanced course on Computer Vision and Pattern Recognition.  It is designed to broaden the knowledge base and enhance the research skills of the students by familiarizing them to the most recent,  most advanced and very active research topics in the field of Computer Vision and Pattern Recognition.    It serves as well the purpose of enhancing the students' scientific communication and presentation skills, and nurturing good research habits.

The course will take a general form of:

1. pre-class paper reading
2. seminar/lecture
3. classroom discussion and project presentation
4. final project and report

In each class we will analyse one or two selected paper form the top-tier Computer Vision conferences (e.g., ICCV, CVPR, ECCV or NIPS).   Students are required to read assigned paper before class.  The class will start with a presentation about the topic of the day, followed by a classroom discussion.  By the end of the course each student is expected to complete a term-project and report based on one selected paper.  This course will present an in depth coverage of recent important topics in computer vision.

In completing this course, students will acquire:

1. A clear understanding of some current and important research issues in computer vision and pattern recognition;
2. The skills and knowledge needed to appreciate research ideas and evaluate good publications in the research field.
3. The ability of conducting effective scientific communications, such as report writing and seminar presentation. 
4. Understand and be able to apply knowledge and skills to project and problems in practice.
5. The skills and habit needed to carry out independent and innovative research.

A working knowledge of Matlab or C/ C++ programming

• Linear algebra
• Calculus
• No prior knowledge of vision is assumed, but the fundamental coursework of ENGN8530 is generally recommended.

No textbook is required, but the following two books can serve as general background reading:

• Trucco and Verri Introductory Techniques for 3 D Computer Vision, Prentice Hall 1998.
• Forsyth and Ponce Computer Vision: A Modern Approach, Prentice Hall 2002.