| Part 0: Course Setup and Logistics |
| Tuesday, 9/8 |
Course overview
|
|
| Thursday, 9/10 |
Python, NumPy, and Matplotlib tutorial
|
|
| Part I: Classical Computer Vision |
| Tuesday, 9/15 |
Cameras and geometric image formation
|
|
| Thursday, 9/17 |
Light, reflectance, color, and sensors
|
|
| Tuesday, 9/22 |
Light, reflectance, color, and sensors II
|
|
| Thursday, 9/24 |
Modeling images
|
|
| Tuesday, 9/29 |
Convolutions and their applications
|
|
| Thursday, 10/1 |
Local features: corner and blob detection
|
|
| Tuesday, 10/6 |
Transformations, alignment, and robust estimation
|
- Lecture slides [pdf]
- Szeliski book, Chapter 8
|
| Thursday, 10/8 |
Optical flow, two-view stereo
|
|
| Tuesday, 10/13 |
Alignment application: multi-view stereo, SfM, image search
|
- Lecture slides [pdf]
- Szeliski book, Chapter 11, 12, 13
|
| Thursday, 10/15 |
Midterm 1: classical vision
|
|
| Part II: Modern Computer Vision |
| Tuesday, 10/20 |
From classical recognition to learned representations
|
|
| Thursday, 10/22 |
Nearest neighbor classifiers, linear classifiers, loss functions, regularization
|
- Lecture slides [pdf]
- Szeliski book, Chapter 5
|
| Tuesday, 10/27 |
Gradients, optimization, neural networks, convolutional layers
|
|
| Thursday, 10/29 |
Convnet architectures
|
|
| Tuesday, 11/3 |
No class — Election Day |
| Thursday, 11/5 |
Image segmentation and object detection
|
|
| Tuesday, 11/10 |
Image segmentation and object detection II
|
|
| Thursday, 11/12 |
Understanding and visualizing convnets
|
|
| Tuesday, 11/17 |
Adversarial examples, texture synthesis, style transfer
|
|
| Thursday, 11/19 |
Recurrent neural networks
|
|
| Tuesday, 11/24 |
No class — Wednesday schedule followed |
| Thursday, 11/26 |
No class — Thanksgiving recess |
| Tuesday, 12/1 |
Attention and transformers
|
|
| Thursday, 12/3 |
Self-supervised learning
|
|
| Tuesday, 12/8 |
Midterm 2: modern vision
|
|
| Thursday, 12/10 |
Generative modeling
|
|
| Tuesday, 12/15 |
Generative modeling
|
|
| Final-exam period |
Project poster/oral presentations |