1 Introduction

1.1 Where to go from our basic building block?

Untitled

1.1.1 Three ‘R’s of Computer Vision

“The classic problems of computational vision:

  1. reconstruction
  2. recognition
  3. (re)organization.”

Jitendra Malik, UC Berkeley

1.1.2 Recognition

Often needs machine learning for compact descriptions of the visual world.

language or vector?

Untitled

1.2 ML for Computer Vision

Learn from and make predictions on data.

  1. A CRASH COURSE for MACHINE LEARNING
    1. We will look at ML as a tool. We will not detail the underpinnings of each learning method.
    2. Please take a machine learning course if you want to know more!

1.3 Data, data, data!

“... invariably, simple models and a lot of data trump more elaborate models based on less data”

— Peter Norvig – “The Unreasonable Effectiveness of Data” (IEEE Intelligent Systems, 2009)

1.4 ImageNet: Large Scale Visual Recognition Challenge

https://arxiv.org/pdf/1409.0575.pdf