Skip to main content

Intro

Intro to Computer Vision​

  • Course details
  • What is computer vision (CV)?
  • Course outline

Basic Image Processing​

  • Image representation
  • Pixel-wise operations
  • Histogram equalization
  • Template matching
  • Morphology operators
  • Connected components
  • Color space

Filtering and Resampling​

  • Noise and filtering
  • Frequency representation
  • Decimation
  • Interpolation

Edge Detection​

  • Intro to edges
  • Basic edge image
  • Edge thinning
    • LoG
    • NMS
  • Edge mask
  • Canny edge detector
  • Other edge related topics
    • Frequency representation
    • Unsharp filter

Curve fitting​

  • Least squares
  • Total least squares
  • RANSAC

Hough transform​

  • Hough transform
    • (m,b)(m,b) parameter space
    • (ρ,θ)(\rho,\theta) parameter space

Image formation​

  • BRDF
  • Pinhole camera
  • Digital camera
  • The human eye

Geometric Transformation​

  • 2D->2D transformations
  • 3D->3D transformations
  • 3D->2D transformations (3D projections)
    • Perspective projection
    • Orthographic projection

Camera calibration​

  • What is camera calibration?
  • Camera extrinsics
  • Perspective projection
  • Camera intrinsics
  • Full camera matrix
  • Calibration methods and distortions

Features​

  • What and why we need features detection?
  • Feature detection
    • Blob detection
    • Harris corner detection
    • SIFT detector
  • Feature description
    • Template matching
    • HOG
    • SIFT descriptor
  • SIFT feature matching
  • Panoramas

Stereo​

  • Structure from motion
  • Triangulation
  • Stereo matching
  • Camera rectification
  • Epipolar geometry
    • Essential matrix
    • Fundamental matrix
    • Estimating the fundamental matrix
  • Other 3D sensors

Neural Networks Basics​

  • The classification problem- again
  • NN history
  • Perceptron
    • Hyperplanes
    • Activation
  • Dense layer
  • Multi-layer perceptron (MLP)
  • Optimization
    • Softmax + cross entropy + loss
    • Gradient descent
  • Basic data preprocessing
    • Data normalization
    • Train, validation and test splits
  • Fully connected net

Convolutional Neural Networks ​

  • ConvNets
    • Convolution layer
    • Pooling layer
  • Overfitting
  • Architectures
    • Alexnet (dropout)
    • VGG
    • ResNet (batch norm)