Computer Vision

Inspiration

This is for the challenge "Explore a Buzzword Topic" of Day 4 of Major League Hacking's "LHD:Build" hackathon.

What it does

It introduces about what computer vision is, what we do in it, what are the challenges and applications of it.

Computer Vision

I explored the buzzword “Computer Vision” and summarized my learnings as given below. Computer vision deals with how computers can gain high-level understanding from digital images or videos. It is named so because the computer tries to understand and automate the tasks that a human visual system can do. For a computer to analyze image data like a human it should be able to describe the contents, summarize or recognize the contents of an image or video it has seen only once. It is a subfield of artificial intelligence and machine learning and is multidisciplinary area of study. It is different from image processing as image processing involves simplifying or enhancing the content of images while computer vision is a type of digital signal processing and is not concerned with understanding content of the image. There are many challenges in the computer vision field, as we still do not know much about how human vision works. In addition, the visual world is very complex and the same object looks different in different lighting, orientation, etc. Some of the various high-level-problems where we have successfully implemented computer vision are- optical character recognition, machine inspection, 3D model building, medical imaging, automotive safety, motion capture, surveillance, merging CGI with actors, fingerprint recognition and biometrics, etc. Some computer vision applications where we try to recognize things in photographs are- object classification (category of object), objet identification (type), object verification (object present or not), object segmentation (finding pixels identifying the object), object recognition (what are the objects and where).

Built With

• chrome