CE 261: Lab #3
View Morphing of Scene from Hoover Tower


Spring 1997

Instructor: Dr. Craig M. Wittenbrink

Office: Applied Science Bldg. #309

HP Phone: (415) 857 2329
UCSC Phone: (408) 459 4099

Due: Thursday May 22. Labs are due at the beginning of class.

All labs will be demonstrated at the start of class, and writeups will be collected at that time. Lab Environment: Computer Engineering/Computer and Information Sciences educational lab. Silicon Graphics Indy's and Indigos.

Location: Applied Sciences 213

Lab Hours: Any time, but you need a keycode

View Morphing.

Kodak Photo CD images are available on the 213 lab machines in directory: /classes/cmpe261/hoover . There are eleven images there. All of the images (except for the first one) are 3072x2048 pixels (width height). These images were collected from the Hoover Tower on the Stanford, Palo Alto campus using a 35 mm camera. The images were digitized onto a Kodak PhotoCD. Images 1 to 11, are in a translation from left to right in front of the window. The window is exactly 8' wide. Image 1, is 1' to the left of the left edge of the window, image 2, is at the left edge of the window, image 3, is 1' to the right of the left edge of the window, etc. All images were taken with the front leg of the tripod 7 feet from the window, at a height of 63 inches above the floor, and a horizontal declination angle of 14 degrees pointing downward. The tripod had a distance from the center to the front leg of 26 1/2 inches.

The lab assignment is to implement View Morphing, as described in [1]. You are to use the image vision libraries to process the images, and to perform the warping. If you want to implement your own interpolation code, I would recommend checking Castleman [3]. The images have been collected with parallel viewpoints, which facilitates not having to do the prewarp.

Each lab is to perform a view morphing between 2 images. Please experiment with using images that are far apart, and ones that are closer together. The goal is to properly morph the far field features of buildings, streets, and trees. So use those as correspondence points as shown in [1].

In Lab 4, you will enhance your lab to perform better with the wrought iron bars. In Lab 3, you may use any means you wish to create your control/ correspondence points for the warp. Each morphed image shall be a blend between at least 2 source images. Additional source images can be used.

Artifacts adaption

To remove the artifacts of the bars, and occlusion effects, there are many approaches. If you wish. You may clarify the artifacts that you discover, and implement methods for dealing with them in Lab 3. One means for extracting the bars may be to threshold, and then perform a connected components analysis. The connected components analysis algorithm is described in Haralick and Shapiro, pages 28-48 [4]. Pseudo code is given for various algorithm variants. The classical algorithm, similar to the algorithm described in Gonzalez and Woods, chapter 2, can be used, Section 2.3.4, or the ``Space-Efficient Two-Pass Algorithm That Uses a Local Equivalence Table'', Section 2.3.5

Lab 3, Lab Write-up

You must show in your write-up the pseudo code of the algorithms used for the view morphing. Demonstrations of the processing on the Hoover tower images will be required. Also describe the implementation of your program. A listing of the program showing that the program compiles, usage etc. is to be included. I recommend that you start on the lab immediately, as programming projects often take longer than expected.

References

[1] View Morphing, by Steven M. Seitz and Charles R. Dyer, in the Proceedings of SIGGRAPH'96, pages 21-30.

[2] ImageVision Library Programming Guide, by Jackie Neider and Eleanor Bassler, Document Number 007-1387-030, Silicon Graphics, 1993.

[3] K. Castleman. Digital Image Processing. Prentice-Hall, 1979. Second Edition 1996, Chapter 8, Geometric operations.

[4] R.M. Haralick, and L.G. Shapiro, Computer and Robot Vision, Volume I, Addison Wesley, 1992.


Copyright, Craig Wittenbrink, 1997.

craig@hpl.hp.com
Last modified Monday, 06-Apr-1998 23:25:22 PDT.