Introduction to probability theory
We have studied probability in several courses through university or elsewhere. In this section, the aim is to fill in the gaps, so that computer vision algorithms that require probability theory can be easily built upon. The motivation to use probability theory in computer vision is to model uncertainty.
What are random variables?
Random variables are used to define the possibilities of an event in terms of real numbers. The values it can represent are random and, by applying certain assumptions, we can restrict it to given range. To get started with random variables, we need to either compute a function that approximates its behavior or assume and prove our hypothesis function through experimentation. These functions are of two types:
- In the discrete domain, random variables' values are discrete. Then the function used to model probabilities is termed as Probability Mass Function (PMF). For example, let be a discrete random variable; its PMF is given by...
