CMSC 426
Image Processing
Spring 2013

Instructor: Yiannis Aloimonos (yiannis at cs dot umd dot edu)
Lecture: Tu-Th 12:30-1:45 - CSI 1122
Office Hours: By appointment or before class - AVW 4475

Important Anouncements !
(Please, reload everytime you check the announcements)

(Jan 22) Welcome to CMSC426

Course Outline

In this class we will cover the following topics:

1. Introduction:
What is Computer Vision? Ongoing Research and Application Areas.
2. Image Formation:
Geometric aspects, Radiometric Aspects, Digital Images, The Human Eye, Camera parameters
3. Filters:
Linar Filters and Convolution, Spatial Frequency and Fourier Transform, Sampling and Aliasing, Noise Reduction small
4. Edge Detection:
Gradient based edge Detectors, Laplacian, Parametric Models
5. Other Image Features:
Hough Transform, Ellipse fitting, Deformable contours
6. Lightness and Color:
Surface Reflectance, Recovering Lightness, The Physics of Color, Human Color Perception, Color Representations
7.Camera Calibration :
Intrinsic Parameters, Extrinsic Parameters
8. Multiple View Geometry:
Stereo, The Correspondence Problem, Epipolar Geometry, 3D Reconstruction
9. Motion:
The Image Motion Field, Estimation of 3D Motion and Structure, Segmentation on the basis of different Motion, Image Compression
10. Shape from Single Image Cues:
Surface Descriptions, Shape from Contours, Shape from Shading, Shape from Texture.

Reading List

There is no required text. We will distribute material from a variety of sources.

Grading

Midterm: 20%, Final: 20%, Homework: 60%

Lecture Notes

Lecture notes will appear as the semester goes on.

Topic 1: Image Formation 1 [pdf] [ppt]

Topic 2: Projective Geometry [ppt] (10 MB)

Topic 3: Linear Algebra Review [ppt]

Topic 4: Camera Calibration [pdf]

Topic 4: Filtering [ppt]

Topic 5: Edge detection [ppt]

Topic 6: Epipolar Geometry [ppt]

Topic 7: Stereopsis [ppt]

Topic 8: Image motion [ppt]

Topic 9&10: Interpretation of image motion fields [ppt]

Homework

Homework 2, due 02/21

Homework 2 is a programming assignment. You are given a set of images, attached here, and you are asked to make a mosaic of the images. This means stiching them together in the same coordinate system, as we discussed in class. The images are taken by a camera rotating around an axis passing through its center. As a result, the images are related to each other via homographies. To estimate those homographies, you should manually correspond four points in the two images that you want to merge.

Deliverables: In one pdf file, include:

a short essay describing your solution process

your code

the final mosaic (you could also include intermediate results)

Helpful materials: We have covered in class all that is needed to solve this problem. For the ones interested, I also attach a classic paper on mosaicing by R. Szeliski
[Images] [Szelinski paper]

Homework 3, due 02/26

The document file containts problems for Homework 3.
[doc] [Image]

Homework 4, due 04/09

Implement the stereo algorithm of your choice, preferably correlation windows. Apply it to one pair of images from the Middlebury website. Show your results.

Homework 5, due ???

Create a video identical to the attached where you replace the text: CMSC 426 Computer Vision with any pattern of your choice. As discussed in class at length, use the features of the letters to compute the homographies relating any two frames the video, so that you know how to render the pattern that you will insert. If you missed class, talk to your classmates or come see the Instructor.
[Video]

Homework 6, due ???

Use calibration images and the 3 images of the block to make a 3D reconstruction.
[Images]

Additional Resources

Online resource of computer vision topics (contains short descriptions and tutorials on basic and advanced topics)

Image Processing Learning resources

Web Accessibility