Lectures slides, notes and annotations will appear in this folder (UMass credentials required).
Links to the Echo360 will be posted on Piazza two weeks after each lecture. For early access submit the request form on Piazza.

Date Lecture Readings Logistics
2/7 Lecture #1 :
Introduction and logistics

Module 1: Image Formation
2/9 Lecture #2 :
Light and color
- Spectral basis of light
- Color perception in the human eye

2/14 Lecture #3 :
Light and color
- Tristimulus theory and color spaces
- Color phenomenon

2/16 Lecture #4 :
Image formation
- Pinhole camera model
- Qualitative properties
  • Szeliski book, Chapter 2

2/21 Lecture #5 :
Image formation
- Cameras with lenses
- Lens phenomenon
  • Szeliski book, Chapter 2

Module 2: Image Processing
2/23 No class (snow day)
2/28 Lecture #6 :
Digital images
- Analog and digital color sensing
- Alignment and demosacing
- Spatial and bightness quantization
- Color displays and colormaps

3/2 Lecture #7 :
Python tutorial

3/7 Lecture #8 :
Image processing: I
- Improving brightness and contrast
- Filtering and convolution
- Smoothing
  • Szeliski book, Chapter 3

3/9 Lecture #9 :
Image processing: II
- Sharpening
- Edge detection
  • Szeliski book, Chapter 3

3/14 No class (Spring break)
3/16 No class (Spring break)
3/21 Lecture #10 :
Corner detection: I
- Local descriptors
- Simple corner detector
- Harris corner detector
- The second moment matrix
  • Szeliski book, Chapter 7

3/23 Lecture #11 :
Corner detection: II
- Local descriptors
- Simple corner detector
- Harris corner detector
- The second moment matrix
  • Szeliski book, Chapter 7

3/28 Midterm review
3/30 Midterm exam (in class)
Module 3: Image Understanding
4/4 Lecture #12 :
Blob detection
- Invariance and equivariance
- Scale equivariant features
- Image pyramid and scale-space

4/6 Lecture #13 :
Feature matching and model fitting
- Szeliski book, Chapter 8 - Feature descriptors
- Matching and ratio test
- Model fitting using RANSAC

4/11 Lecture #14 :
Transformations and alignment
- Image transformations
- Transformations from matching
- Image warping
  • Szeliski book, Chapter 8

4/13 Lecture #15 :
Recognition by alignment
- Instance matching
- Vocabulary trees
- Low distortion correspondences
- Structure from motion

4/20 Lecture #16 :
Introduction to recognition
- What is recognition?
- Brief history of recognition
- Current trends
  • Szeliski book, Chapter 6

4/25 Lecture #17 :
Representations for recognition
- The machine learning framework
- Role of representations
- A tale of two representations (HOG and BOVW)

4/27 Lecture #18 :
Machine learning
- Learning framework
- Decision trees
- Nearest neighbor classifiers
  • Szeliski book, Chapter 5

5/2 Lecture #19 :
Machine learning
- Perceptrons
- Linear classifiers and optimization
  • Szeliski book, Chapter 5

5/4 Lecture #20 :
Object detection
- Detection task
- Datasets and evaluation metrics
- Sliding window detector
  • Szeliski book, Chapter 6

5/9 Lecture #21 :
Object detection; Deep learnining
- Region-based detector
- Properties of two layer networks
  • Szeliski book, Chapter 6

5/11 Lecture #22 :
Deep learning
- Learning via backpropagation
- LeNet and AlexNet
- Visualizing filters in AlexNet
  • Szeliski book, Chapter 5

5/16 Final review
5/22 Final exam (1:00 PM - 3:00 PM, LGRC A301)