Компьютерное зрение

Лекторы: 

Программа курса

  1. Практикум
  2. Image processing 8
    1. Image formation 1
      • Human Vision System
      • CCD Camera (Lens, Matrix), its optical parameters (ISO, F, exposure)
      • Color spaces
    2. Filtering 3
      • Global filtering (contrast, brightness, gamma correction)
      • Noise types & origin
      • Image domain linear filtering
        • Averaging
        • Gradient (1,2,..)
      • Frequency domain
        • Convolution & deconvolution
        • Hi-pass, low-pass, modifications
        • BM3D
      • Super resolution
        • One image (upsampling)
        • Multi image with motion information
        • Video
        • Algorithm review (Frequency-domain, interpolation based, regularization based, learning based)
      • Deblurring
  3. Segmentation 1
  4. Compression 1
    1. Image compression (JPEG)
    2. Video compression (MPEG)
  5. Feature detection 2
    1. Points
      • Corner detection
      • Descriptors: SIFT, SURF, HOG, Haar
    2. Edges
    3. Lines
  6. Geometrical Vision (6)
    1. Single view geometry
      • Projective camera
      • Distortion
      • Homography
    2. Two view geometry
      • Fundamental matrix & Essential matrix
      • Estimation methods
      • Outlier detection: RANSAC
    3. Multiple view geometry
      • 2d-3d correspondence
      • Bundle adjustment
    4. Stereo
    5. Range data
      • Acquisition devices
      • Filtering
      • ICP
  7. Recognition (14)
    1. Introduction to ML
      • Principles of learning
      • PCA
      • SVM
      • Neural networks
    2. Face detection & recognition
    3. OCR
    4. Advanced topics in recognition (multi-part recognition)
      После GV – коллоквиум (п.)