Published in News

Vectors instead of pixels

by on21 December 2012 2333 times
pic

If you believe the predictions of Philip Willis and John Patterson from the University of Bath in England, the age of pixels, at least in video compression algorithms, is coming to an end. Although the basic unit of raster images that the video consists of are still pixels, it has been known for a long time that there are  limitations that arise from them. For example, scaling to different, especially higher resolution, makes image quality turn to crap and increases perceptual artifacts with no gain in quality of information.

Video scaled from smaller to greater resolution will still have the same number of distinctive elements as before and in contrast to what we see in the CSI series, raster video registration increase for 10 or 100 times does not make 10 or 100 times more data a computer can register or read. Also, larger grids necessarily take more computer memory and bandwidth which slows systems performance.

The solution is to replace raster codec with vecotor video codecs, and exactly that is being developed at the University of Bath. The technology named Vectorised Streaming Video (VSV) first converts raster to vector, which is then successfully scaled to any resolution without loss of information. The algorithm is apparently similar to those for autovectorization you have in Adobe Live Trace. Willis and Patterson also claim they have solved the problems of colorization of vector spaces between the geometric primitives and that now the result of vectorization is fotorealistic as well as the raster version. The first functional VSV codec supposedly should be on the market in three to six months.

What do you think about this new technology? Tell us in our Forum

Last modified on 21 December 2012
Rate this item
(0 votes)

Leave a comment

Make sure you enter all the required information, indicated by an asterisk (*). HTML code is not allowed.