A project-based learning approach to teaching computer vision at the undergraduate level

Abstract

This paper documents an attempt to use the project-based learning approach to develop and teach a computer vision class in the undergraduate electrical and computer engineering program. The project-based learning approach is one of the innovative methods promoted in the Advanced National Effective Teaching Institute (NETI-2). It combines the traditional lecture-based deductive learning with the inductive learning described by Prince & Felder in a 2006 JEE paper. The paper describes the class objectives as well as the projects that were developed to achieve those objectives. The paper also discusses the challenges encountered during both the development and the delivery phases of the class. These challenges include, identifying suitable class projects, achieving adequate student assessment, and most importantly, students' resistance to the new learning paradigm. The paper also includes assessment results including the faculty assessment reports and the student evaluations of the course.