Introduction to Computer Vision

Computer vision is the problem of getting a machine to recover meaning from pixels, and this graduate course walks you up that ladder from raw image formation through filtering, edges, texture, and segmentation to the higher-level representations that feed into classification and recognition. Expect a mix of mathematical work and implementation: you will code the classical pipeline yourself, read papers, and build toward a team project, with later weeks pivoting to case studies in deep learning. It assumes comfort with linear algebra, signals, and probability, and serves as the foundation graduate students lean on before moving into more specialized work in recognition, 3D vision, or learning-based perception.