Computer Vision (often referee to simply as CV) can loosely be thought of as techniques that allow a computer system to understand specific qualities about visual data and/or spatial data. This visual data could be a static photograph, movie file, or live webcam feed.
Some qualities that we may want a computer to understand about visual data include;
CV is often described as a sub-field of machine learning, as more complex tasks typically require the use of machine learning techniques and models to determine what is occurring in visual data.
It is possible to determine qualities or to extract feature data of visual data through simpler image processing techniques. Some people do not consider this latter set of techniques as CV, and instead prefer to call them “image processing”. However, for the purposes of this course, if the goal of any processing is to understand something about the visua data, we will consider it CV.
Please read the following post to understand more about CV:
The following TED talk does a nice job overviewing some of the more interesting problems in CV, discussing its relation to machine learning, and discussing techniques that are often employed.