Intelligent Resizing

Everyone seems to be totally fascinated about the intelligent image resizing algorythm developed by Shai Avidan and Ariel Shamir. Me too, i see ;). Today afternoon i have seen the first samples written in Java and Flash AS3, they all show the manner, but i think this is not the end of the line.

So, i downloaded the White Paper (~20Mb) and got the idea. Bascially it’s based upon a physcial princip, called energy conservation. But now the ingenious part, seams(the vertical or horizontal lines that you can see during resizing) are added or removed without destruct the basic image data. Original data is availabe as long the image is saved without a lossless compression. It’s ingenious, isn’t it?

After nearly 20 years of multimedia it seems, that physic principals will give us the possibility to “understand” pictures/photos in a way that there could be a programm/algo that tell you what is inside a picture. What comes up next? Newton…

“Attention, fallin tree on street in 2 seconds. This was a live report from your guardian angel aka. CCTV-cam.”

As everbody could see inside the youtube-video, there is also an easy way to define your own croppable areas, think about the possibilities, doing a photo correction in 1min instead of 1h.

we will see.

PS.: I am writing in english again, because i wanto to be listed at the macromedia news aggregator, but i think this shouldn’t be a problem.