M W 10:00 AM -11:30 AM and by appointment
This is a graduate-level/advanced undergraduate seminar exploring the algorithms and state- of-the-art techniques in the field of computer vision and image processing. Topics covered in the first half of the seminar include image processing, pattern recognition, feature detection, image and object segmentation, computational photography, object detection, and face recognition. In the second half of the semester, students will read and lead discussions on both classic and recent literature in computer vision, machine learning, and related fields. A familiarity with linear algebra and statistical methods is recommended but not required.
*lecture notes come from a variety of sources including Mubarak Shah UCF, Robert Collins PSU, Duda et al Pattern Recognition, Victor Lavrenko, Richard Szeliski Computer Vision, Fei Fei Li Stanford, etc.
|Date||Class Summary||Examples and Code||Homework|
|8/24/2016||Introduction to what is computer vision.||Read Chapters 1 in Computer Vision.|
|8/31/2016||Linear Algebra refresher. Image formation and color spaces.||[slides] . Matlab handout . Matlab book chapter . Linear algebra in 4 pages||Read Chapters 2.1.1 - 2D points, 2D lines, 3D pints, 3D planes.
2.1.2 - 2D transformations, 3D transformations.
3.1, 3.2 Point operations and Linear filtering in Computer vision.
. Read the Color transfer paper. Color Transfer paper
|9/7/2016||Matlab Tutorial. Image formation, Image filtering||[slides]||Read Chapters 3.3 - More neighborhood operators Homework 1 . Image 1. Image 2. Project 1 . Color image 1 . Color image 2|
|9/14/2016||Edges, Derivative of Gaussian, Laplacian, Sobel, Prewitt, Canny, Seam Carving||[slides]||Read 4.1 and 4.2 in the book. Read the seam carving paper Seam Carving . Homework 2 . Image 1. Image 2.|
|9/21/2016||Interest points, eigenvalue/eigenvector, Harris corners||[slides]||Read 4.1 in the book.|
|9/28/2016||SIFT Interest points, Hessian, image descriptors, image pyramids, difference of gaussians, Scale Space||[slides]||Review 4.1 in the book. Read the paper on SIFT. Homework 3 . Puzzle . Piece 1 . Piece 2|
|10/5/2016||Projections, covariance, dimensionality reduction, PCA, LDA, eigenfaces||[slides]||Read 14.2.1 in the book. Read the paper on Eigenfaces. Project 2|
|10/26/2016||Classification, linear regression, Boosting||[slides]||Lab 4. Lab5 . Presentation instructions|
|11/2/2016||Classification, Bag of Visual Words, Random Forest , Neural Networks||Batches Meta . Data30k.mat . Labels30k.mat. Mdl.mat. [slides]||Lab 6 . Final Project Description|
|11/9/2016||Introduction to Deep Learning, Backpropogation, Sigmoid, ReLU||[slides] . Deep Learning Matlab Demo||Extra Credit . Captchas . Numbers|
|11/16/2016||Hidden Layers, Softmax, Convolutional Neural Networks, Pooling, Autoencoders, Fine-tuning, transfer learning, multimodal learning, RNN, LSTMs||[slides] . Using CNNs as a feature extractor||Lab 8|
|11/23/2016||No Class - Thanksgiving|
|11/30/2016||Research paper presentations|
|12/7/2016||No in-class, final project workday|
|12/14/2016||Final Project Presentations||Final Project presentations will be in the ICE center starting after the CSC Senior Project presentations.|