Lecture I: Introduction to Computer Vision

Image

• 2D array of numbers
• Gray levels 0~255
• Color image is 3 2D array of numbers

Video

• Sequence of frames

Image Formation

• Light source
• Camera
• Scene

Perspective Projection

• Pictures are 2D array of numbers describe 3D real world, and that is one of the hardness of computer vision.
• Pin hole camera model
• 

Applications

• Human detection
• Object recognition
• Face recognition
• Facial expressions - detecting driver alertness from eyes' moving
• Lipreading
• Video surveillance and monitoring - obeject detecting -> object tracking -> object categorization and classification -> event or activities recognition
• Unmanned aerial vehicles (drones)

Dataset Introduction

• Almost all human action dataset, review here if need.

Lecture II: Filtering

Basic Image Knowledge

• Image kind
  • Binary - only 0 and 1
  • Gray Scale - only 0~255
  • Color - 3D 0~255
• Image noise
  • Light variations
  • Camera electronics
  • Surface relectance
  • Lens

Correlation and Convolution

• Convolution adds two flip operations comparing to correlation

MATLAB funcs about filter

refer to: here

Author：Hongyang Wang, http://mr-why.com
Feel free to repost if obey Creative Commons BY-NC-ND 3.0