Images and videos are omnipresent today, on social media, space missions, medical and physical sciences or collected from security cameras. Analysis of large volumes of data stored in images and videos is needed to search for features and patterns in order to extract useful information from them. This requires knowledge of both visualization and data science techniques to make sense of the images. The algorithms developed for this purpose are essential for applications in multiple fields, including: autonomous driving, national security, disaster response, urban planning, and personal communications, among many others.

This course will cover the basis of computer vision techniques used to extract information from large volumes of 2D imaging data. The syllabus contains relationships between the 3D world and 2D images, visual tracking, visual recognition of objects and events, higher-level reasoning for scene understanding, projective geometry, modeling and calibrating cameras, geometric primitives and their uncertainty, stereo vision, motion analysis and approximating three-dimensional data. By the end of this course, students will be able to take an image and write algorithm exploiting advanced data science techniques to extract maximum information.