Monday, February 23, 2009

UIST papers: "Annotating Gigapixel Images" and "Foldable Interactive Displays"


Annotating Gigapixel Images:

This paper is basically about a group that has devised a way to go 
about and label these gigapixel images, which are extremely large 
images commonly used for geographic photos of the earth, label 
them with text or sound. The labels can denote regions, cities, 
or even something as small as a person and label that. Then it 
goes on to discuss how these labels should be displayed on the 
screen when panning and zooming. It details these equations for 
the different renderings in the paper.








Foldable Interactive Displays:

This paper was about a research team trying to come up with a way to
have cost effective and small space consuming displays. Essentially,
foldable projection screens that can move around. The title of the paper
is a little
deceiving, you think it would be a kind of screen that you 
could fold up into your pocket and then take it out and use it anywhere. 
But this is just a projection screen that can fold up, you have to have 
a projector to put the images on it. Basically, the have 4 different types 
of foldable screens: a fan, an umbrella, a newspaper, and a scroll. And 
they put Rf emitters into these so that a projector could track where the 
screen was as you carry it around. You can also do things like flipping 
over your screen and the display acts accordingly by flipping your 
"paper" that your reading over. Also you can resize the screen, for 
instance, on the newspaper you can fold it up some and the display 
resizes accordingly. By adding additional LEDs into the display input 
devices can be used like a stylus to track user operations. Overall, this 
paper underwhelmed me because its seems very impractical to have 
projectors everywhere, and that's what it needs for this to even be 
useful. So the research seems pointless to me.


1 comment:

Josh Myers said...

This is a really cool technology. It seems like everyone is doing some sort of projection-style computing. Being able to take the object and physically manipulate it right in front of you certainly has its advantages. Just projecting an image on an object is one thing, but I would prefer an object that has the screen built in, so that it is more portable.