Representation and Modeling for Image Analysis
MIT OpenCourseWare seminar on the two core components of computer vision and image analysis algorithms: representation and modeling. Lectures cover representations such as contours, level sets, and deformation fields, paired with estimation techniques including manifold fitting, Markov random fields, expectation maximization, and clustering. Each mathematical concept is followed by discussion of two or three research papers applying it in computer vision, medical imaging, and biological imaging. The course aims to build both the mathematical foundations and the judgment to recognize when a given technique will improve an analysis. Materials include lecture notes and paper readings from MIT's Electrical Engineering and Computer Science department, freely available through MIT OpenCourseWare with no certificate offered.