Abstract: Computer vision is part of artificial intelligence (AI) that focuses on images or video. Its goal to make computers “see” the world, like our eyes or beyond. It is useful for many applications, and recently is used a lot in self-driving cars, surveillance, mobile phones, and industry (such as Amazon Go, Google Image, Facebook Vision Tag, etc). In this talk, I discuss the general four pillars of computer vision, and how machine learning, which includes deep learning, becomes an important approach to solve vision problem. Particularly, I introduce my work on bad weather visibility enhancement using deep learning. My talk targets a general audience who does not have much knowledge in computer science.