Week 1 (Intro, Image formation)
Week 1 (Intro, Image formation)
Introduction (#1)
Image formation, part 1 (#2)
Sign up on Piazza, Gradescope
Homework 1, Mini-project 1
Richard Szeliski book (RS), Chapter 1 and 2
The speed of processing in the human visual system, Thorpe et al., Letters to Nature, 1996
Linear algebra review ++ (via David Kriegman)
Random variables review (via David Kriegman)
Image formation, part 2 (#3)
Light and color, part 1 (#4)
Homework 1, Mini-project 1
Image formation (Szeliski Book Ch. 2.1)
Light and Color (Szeliski Book Ch. 2.2 & 2.3)
Light and color, part 2 (#5)
Digital images (#6)
Homework 3, Mini-project 1
RS 2.3
Review lecture (#7)
We talked about some topics from the survey, color phenomenon, and some research
Image processing, part 1 (#8)
Contrast enhancement
Image filtering, intro
Homework 4, Mini-project 2
https://en.wikipedia.org/wiki/Computational_photography (Also Ch. 10 in the RS book)
https://www.ru.nl/astrophysics/black-hole/imaging-a-black-hole/
https://ai.googleblog.com/2018/11/night-sight-seeing-in-dark-on-pixel.html
Image processing, part 2 (#9)
Gaussian filter, median filter, types of image noise, sharpening
Image processing, part 3; Corner detection, part 1 (#10)
Wrap up image filtering: edge detection
Get started on corner detection --- role of features for image matching, designing a simple corner detector
Homework 4/5, Mini-project 2
Corner detection, part 2 (#11)
Harris corner detector
Blob detection (#12)
Homework 5/6, Mini-project 3
Feature descriptors for matching (#13)
RANSAC for model fitting (#13)
Image transformations (#14)
Application: Panoramic image stitching (#14)
Homework 6/7, Mini-project 3
Review #2 (#15)
A brief history and overview of visual recognition (#16)
Homework 7, Mini-project 4
Recognition by alignment #17
scaling instance matching
local features for shape matching
non-rigid transformations
Learning to recognize #18
The ML framework
Image representations -- HoG and Bag of Visual Words
Homework 8, Mini-project 4
Chapters 6, 7 and 9 from Richard Szeliski’s book.
Shape matching and object recognition using shape contexts, Belongie, Malik and Puzicha, PAMI 2002 (paper)
Shape matching and object recognition using low distortion correspondences, A.C. Berg, T.L. Berg, J. Malik, CVPR 2005 (paper)
Demos: Video google, Oxford building search, Sculpture retrieval
https://en.wikipedia.org/wiki/Histogram_of_oriented_gradients
Learning to recognize, part 2 (#18)
The ML framework
Image representations -- HoG and Bag of Visual Words
Machine learning, part 1 (#19, #20)
Decision trees
Nearest neighbor classifier
Preceptrons and linear classifiers
Learning as optimization
Homework 9, Mini-project 5
Chapters 6, 7 and 9 from Richard Szeliski’s book.
Decision tree learning and material are based on CIML book by Hal Daume III (http://ciml.info/dl/v0_9/ciml-v0_9-ch01.pdf)
Figures for random forest classifier on MNIST — Amit, Geman and Wilder, PAMI 1997 — http://www.cs.berkeley.edu/~malik/cs294/amitgemanwilder97.pdf
Machine learning, part 2 (#21)
Decision trees
Nearest neighbor classifier
Perceptrons and linear classifiers
Learning as optimization
Object detection, part 1 & 2 (#21, 22)
Detection = repeated classification
Sliding window detectors
Region-based detectors
Datasets and benchmarks (PASCAL VOC, INRIA Pedestrians, MSCOCO)
Homework 9, Mini-project 5
No class on Tuesday due to Wednesday schedule
Deep Learning (#22)
From perceptrons to multi-layer networks
Learning and optimization
Convolutional networks
Homework 9, Mini-project 6