Instructor: Thomas Serre
Course: CLPS1520: Computational Vision
A detailed introduction to computational models of biological and machine vision summarizing traditional approaches and providing experience with state-of-the-art methods. Topics include fundamentals of image processing, visual perception (surfaces, color, depth, texture and motion) as well as object recognition and scene understanding. Connections to contemporary research in computer vision and computational neuroscience will be emphasized highlighting how computational models may motivate the development of new hypothesis for experiment design in cognitive psychology.
What students said about the course:
"I felt that the entire class came away from the course with valuable and first-hand knowledge."
"Despite the presence of students studying pure neuroscience, pure computer science, graduate students and even freshmen, no one was left behind."
"Completing MATLAB projects in this course was one of the most satisfying experiences I've had at Brown."
Instructional Technologist's Comments:
Professor Serre uses standard research data tool MATLAB to teach computer science. This is a novel method for teaching this course and is an innovative use of MATLAB. A significant amount of work went into the redesign of this course by Prof Serre, a new faculty member at Brown. Student work and feedback provides evidence that this approach is effective and that students with little or no computer programming skills can complete course work and learn the fundamentals of computer visualization. The course in its entirety is extraordinarily designed. Professor Serre restructured the interdisciplinary course to bring together students from diverse backgrounds (in computer science and neuroscience) to learn about computational vision in a hands-on course. His transformation used MATLAB, a programming language frequently employed by researchers, to allow students to recreate seminar results in the field. These programs allowed the students to understand the mechanisms by which the processes discussed in the class were carried out, rather than just memorizing what occurs. He build in a range of tasks, from conceptual-level material to bonus questions probing more advanced programming skills in order to allow students new to computation skills to have a sense of accomplishment, while providing a sense of pride to those who challenged themselves with the more advanced problems.