|Part I: Representing images as 2D arrays of pixels|
|Wed, Jan 8||Introduction; The Camera Response Function||Digital images; computing camera response functions from images. Slides: [Printer friendly] [Color]||Sections 1.1-1.2, 2.1, 2.2, 2.4.2 (only paragraph entitled "silicon sensors"), 2.6.2 from Castleman book; Sections 1 and 2, up to Eq (2), from paper. As you read the paper, note that film response curve and camera response curve in the case of digital cameras, are one and the same.||The HDRShop home page.
Rendering with Natural Light (a movie that uses high-dynamic-range photography to capture outtdoor illumination and re-use it for image synthesis)
|No tutorial during week 1.|
|Fri, Jan 10||Assignment||Assignment 1 is out, due January 27.|
|Wed, Jan 15||Pixel components: color and alpha||Color image acquisition; alpha matting and the matting equation Slides: [pdf]|
|Tutorial||Matting and linear equations|
|Part II: Representing images as continuous 1D and 2D functions|
|Wed, Jan 22||Computing 1D image derivatives||Least-squares polynomial fitting; intensity derivatives; weighted least squares; RANSAC Slides: [pdf]||Polynomial fitting demo to be shown in next week's tutorial: polydemo.zip.
If you download it, run it for a variety of fits (LS, WLS, 1st degree, 2nd degree, etc) to see their effect.
|Tutorial||Questions about Assignment 1|
|Mon, Jan 27||Assignment||Assignment 1 due at 11:59pm.||Worth 10% of the final mark.|
|Wed, Jan 29||Representing 2D image curves||Local analysis of curves: the tangent & normal vectors, the moving frame Slides: [pdf]||To run the demo shown in class, unpack the zipfile curvedemo.zip |
You should run the demo for a variety of fits (LS, WLS, 1st degree, 2nd degree, etc) to see their effect on the estimated curve.
|Tutorial||Polynomial fitting and 2D curves||Tutorial notes on polynomial fitting and 2D curves by Micha Livne.|
|Wed, Feb 5||Edge detection||Local analysis of 1D and 2D image patches: the Image Gradient. Case study: Painterly Rendering Slides: [pdf]||Paper by Litwinowicz on painterly rendering (this is not required reading).|
|Tutorial||Description of A2, Questions and answers about Assignment 1, part B [slides], and Paper on Accidental Pinhole and Pinspeck cameras (this is not required reading).|
|Fri, Feb 7||Assignment||Assignment 2 is out, due March 2 (two days later than originally advertised!).|
|Wed, Feb 12||Corner detection, Intelligent Scissors||Relation between local shape near extrema and the eigenvectors/eigenvalues of the Hessian; relation between eigenvalues & the trace & determinant of a matrix; localizing edges as zero crossings of the Laplacian; the Lowe feature detector: finding non-cylindrical points through eigenvalue analysis of the Hessian of the Laplacian Slides: [pdf]||Paper by Mortensen on Intelligent Scissors (this is not required reading). |
This paper and the technique will be covered in one of the forthcoming tutorials.
|Tutorial||Questions about Assignment 2 and a paper on Object Recognition by Pedro F. Felzenszwalb. [slides]|
|Wed, Feb 19||No lecture||Reading week|
|Part III: Multi-resolution image representations|
|Wed, Feb 26||Template matching and correlation||Representing images as vectors; evaluating similarity using RMS distance error, cross-correlation and normalized cross-correlation; Slides: [pdf]||To run the demo shown in class, unpack the zipfile corrdemo.zip |
and type corrdemo at the matlab prompt.
|Sun, Mar 02||Assignment||Assignment 2 due at 11:59pm.||Worth 10% of the final mark.|
|Wed, Mar 5||Principal Component Analysis||Face recognition using Eigenfaces
|Section 13.6 from Castleman||To run the demo shown in class, unpack the zipfile recognition_demo.zip
and type pca_recdemo at the matlab prompt.
|Midterm||BA 1190, 6 pm||Worth 20% of the final mark|
|Wed, Mar 5||Assignment||Assignment 3 is out, due March 19.|
|Part IV: Alternative image representations|
|Wed, Mar 12||Gaussian Pyramids||The idea of representing an image at multiple resolutions is described and studied using Gaussian and Laplacian Pyramids.
|Original paper by Burt and Adelson on the Gauss/Laplacian pyramids. You should read up to, but not including, section entitled Entropy.||This Matlab demo shows how a 1D image changes as we smooth it with a sequence of Gaussians of increasing standard deviation|
|Wed, Mar 19||The Haar Wavelet Transform||Wavelet compression of 1D and 2D images Slides: [pdf]||A tutorial paper on the Haar Wavelets.||Matlab wavelets demo shown in class.|
|Wed, Mar 19||Assignment||Assignment 3 due at 11:59pm.||Worth 10% of the final mark.|
|Wed, Mar 19||Assignment||Assignment 4 is out, due April 2.|
|Tutorial||PCA, a few theoretical details.|
|Wed, Mar 26||Polynomial fitting vs. correlation; Matching images using SIFT||Analysis of WLS polynomial fitting and image smoothing as a template matching operation; template matching expressed as a multiplication of an image with a Toeplitz matrix; Gaussian image smoothing; SIFT-based feature detection; the SIFT descriptor; image matching using SIFT Slides: [pdf]||Sections 1-3 of paper by David Lowe describing SIFT.||Web page on SIFT (with demo code)|
|Part V: Introduction to 2D Image Transformations|
|Wed, Apr 2||Homogeneous coordinates||Homography-based image warping Slides: [pdf]|
|Wed, Apr 4||Assignment||Assignment 4 due at 11:59pm.||Worth 10% of the final mark.|
|Apr 9-30||Final Exam||Location, time and date to be announced||Worth 40% of the final mark|