CS231N, Lecture 1 | Introduction to Convolutional Neural Networks for Visual Recognition

 

안녕하세요, 

오늘은 굉장히 유명한 Stanford 대학교의 CS231N 강의에 대해 리뷰해보도록 하겠습니다.

 

인공지능을 하시는 분들이라면 해당 강의를 한 번씩은 접해보셨을 거라 생각합니다.

한번 더 공부하시고 싶은 분들이나 처음 접하시는 분들에게 도움이 되는 글이길 바라면서

Lecture 1 | Introduction to Convolutional Neural Networks for Visual Recognition  시작하도록 하겠습니다.

 

---

 

1. 소개

The amount of visual data in our world has really exploded to a ridiculous degree in the last couple of years.

And, this is largely a result of the large number of sensors in the world.

Probably most of us in this room are carrying around smartphones, and each smartphone has one, two, or maybe even three cameras on it.

 

the field of computer vision is an interdisciplinary field, and it touches on many different areas of science and engineering and technology.

interdisciplinary field

2. 아젠다

- A brief history of computer vision

지구의 역사에 대해 설명

- CS231n overview

CS231n focuses on one of the most important problems of visual recognition - image classification.

Image classification is super useful on its own and could be applied all over the place for many different applications.

 

 

3. Computer vision

There is a number of visual recognition problems that are related to image classification, such as object detection, image captioning .

Object detection

* Object detection : 객체탐지, 이미지나 동영상에서 사람, 동물, 차량 등 의미있는 객체의 종류와 그 위치를 정확하게 찾기 위한 컴퓨터비전 기술

Rather than classifying an entire image as a cat or a dog or a horse or whatnot, instead we want to go in and draw bounding boxes and say that there is a dof here, and a cat here, and a car over in the background, and draw these boxes describing where objects are in the image.

 

* Image captioning : 쉽게 말해 AI가 이미지를 읽고 이해하며 텍스트로 설명하는 기술

where given an image the system now needs to produce a natural language sentence describing the image. It sounds like a really hard, complicated, and different problem, but we`ll see that many of the tools that we develop in service of image classification will be reused in these other problems as well.

 

* Convolutional Neural Networks(CNN) have become an important tool for object recognition.

* Why have they only suddenly become popular in the last couple of years?

- there`s a couple really key innovations that happened that have changed since the '90s.

1. Computation : researchers can explore with larger architectures and larger models, and in some cases, just increasing the model size.

2. data ( between now and the '90s data ) : algorithms are very hungry for data.

 

 

Computer vision can be very useful.

It can go out and make the world a better place in various ways (self-driving cars and robotics etc..)

 

Pre-requisite

• proficiency in Python
• high-level familiarity with C/C++
• College Calculus, Linear Algebra
• Machine Learning