Re:Fractal compression for generating resolution-independent...
by neckro23 on Friday December 05, 2014 @06:47PM
(#8707634)
I wondered what had happened to it. In the early 90s it seemed like it would be the future of image compression, and here we are in 2014 still using DCTs.
[ Parent
| on HN ]
Re:Fractal compression for generating resolution-independent...
by ska on Saturday December 06, 2014 @12:52AM
(#8708501)
The real stumbling block is the difficulty in generating anything embedable, and the need to iterate on the decoder side. You can get pretty good results for general images, but not "better enough" to justify this cost over, say as wavelet threshold plus entropy coding. You also can have artifacts of the shape Of the domain support, and getting rid of these makes the encoder quite expensive ( less of a problem)
Re:Fractal compression for generating resolution-independent...
by krylon on Saturday December 06, 2014 @05:08AM
(#8708857)
A couple of years back I saw a documentary on fractals ("Fractals: The Colors of Infinity", narrated by Arthur C. Clarke, with music by Dave Gilmour of Pink Floyd - that was one groovy documentary!) where they showed this mathematician who apparently had a research grant from some military-intelligence-ish government agency to develop fractal image compression for their spy satellites. I wondered why I never that technology in action, and I figured it had something to do with the amount of secrecy surrounding spy satellites.
It is kind of sad that the reason is so much more mundane.