We suspect that if you give a computer an image of a part of some object, it would be helpful in analyzing that image if the computer can imagine what surrounds the field of view of the image. We haven't had time here yet to discuss the algorithms or the applications in detail, but we can show what may be some of the first images of such "imagined surroundings", in this case for some one dimensional images.
In these figures, the image shown to the computer is in the center two quarters of the field of view, and is represented with a lighter shaded curve. The computer's recollection or model of the image, as well as the surroundings imagined for the image on the left and right, are shown with circles and a darker connecting line. Note that what kinds of stuff the computer "imagines" to be in the surroundings seem related to what the computer sees in the image. We are very curious to see if this trend continues for 2D images, and how far we can go in making imagination algorithms increasingly sophisticated! Check in here periodically for updates, as well as details, in days ahead.
You might have noted that in addition to having a wierd imagination, the computer's impression of the image it was given to work with was also less than perfect. We think that will improve in days ahead, as well.
Send comments, possible answers to problems posed, and/or complaints, to philf@newton.umsl.edu Note: This page contains original material, so if you choose to echo in your work, in print, or on the web, a citation would be cool.
(Thanks. /philf :)