Maseeh Engineers Explain: Computer Vision
Atul Ingle, Computer Science Assistant Professor: Computer vision is a subdomain within computer science whose goal is to design automated systems that can see and perceive the world around us the way humans do. There is a real need right now for AI systems to interact with the real, physical world, to move beyond just text. And to enable these AI systems to interact with the real world, they need eyes, so to speak. Right?
Seeing is something that comes extremely naturally to us human beings, but if you spend even just a few minutes trying to think through how complicated a simple task such as, say, looking at the surface of this table and realizing that it's actually flat and at a certain angle with respect to your viewpoint—These kinds of things that humans are able to do so naturally are extremely hard to encode into an algorithm.
Imagine industrial applications where you have robots assembling small, little parts in, say, an assembly line for a far, or it may be a robot that's doing a surgical procedure; it may be a drone that's doing a really fast flyby. So, there are lots of different applications where the ability to see as humans see—Encoding that into a computer algorithm is essential. And that's really what computer vision is all about.
Let knowledge serve. Maseeh College of Engineering and Computer Science at Portland State University. pdx.edu/maseeh.