CSCI 8820 - Computer Vision and Pattern Recognition
Instructor: Dr. Suchi Bhandarkar
Fall 2002

Text: Machine Vision, by R. Jain, R. Kasturi and B.G. Schunck, McGraw-Hill Inc., 1995.

Course Description: The course will cover fundamental issues and paradigms dealing with computer vision and pattern recognition. The material covered in the course will be divided into three phases: low-level vision, intermediate-level vision and high-level or cognitive vision. The material on low-level and intermediate-level vision will constitute roughly Chapters 1--6 of the textbook. The material on high-level vision will be based on selected sections from Chapters 7, 9, 10, 11, 14 and 15. In addition, if time permits and depending on the interest of the class, specialized topics in computer vision and pattern recognition would be considered. These include: parallel algorithms and parallel computer architectures for computer vision and pattern recognition, expert systems for image interpretation, visual information systems and neural computing for computer vision and pattern recognition. A topical list is as follows:

• Introduction to Computer Vision and Pattern Recognition.

• Image processing basics.

• Region analysis.

• Edge detection.

• Image segmentation

• Feature Extraction

• Techniques for extracting 3D shape information from 2D images

• Texture analysis.

• Motion analysis.

• Object recognition

This course is going to be more along the lines of a graduate research seminar than a regular classroom instruction-based course. Being a special topics course, it will be primarily targeted towards an audience that is either already doing research in computer vision, image processing or some related area or seriously considering doing research in such an area. In addition to these topics covered from the textbook, there will be papers assigned from the research literature. Students will be expected to read these papers and the material therein will be discussed in class and may also appear in the examinations and homework assignments. The students are expected to do a fair amount of independent study and show initiative and serious commitment towards learning about various concepts and research problems in computer vision and pattern recognition.

Class Meeting Times: This course will be organized as a combination of formal classroom instruction and a student seminar. The class will meet on Tuesdays and Thursdays during the class period. Since this is a 4 credit hour course, the students are expected to devote the remaining class period and an additional 6--8 hours every week towards a self-study project of their choice. On October 17, Thursday each student or team is expected to submit a formal project proposal. The students will be expected to present their work towards the end of the semester and also submit a technical report/term paper, which will be due on December 12, Thursday, 2002 by 12 noon.

Prerequisites: The primary prerequisite is a strong commitment to learning and hard work. Competence in programming in some high-level language (C, C++, Java etc.) will be assumed. Prior exposure to programming in computer graphics or image processing is a strong plus. The course will be mathematically involved in certain parts. Strong mathematical and analytical skills are a definite advantage.

Grading:

Test 1: 20%
Test 2: 20%
Homework assignments: 20%
Project Report and Presentation: 40%

The tests will be held during the class period and the test dates will be announced at least a week in advance. Homework assignments will be due at the end of the class period on the specified day.

Students are expected to familiarize themselves and comply with the Computer Science Departmental Policy Statement on Academic Honesty (attached).