3 ms·
At one point the holy grail of compression algorithms was fractal encoding which if I understand correctly is basically asking the question what fractals do I n
by somat 4y ago
At one point the holy grail of compression algorithms was fractal encoding which if I understand correctly is basically asking the question what fractals do I need to get my image. theoretical compression ratios off the chart, the problem was the only method found to solve the problem was the "grad student algorithm", that is lock a grad student in a room with plenty of pizza and coffee until they figure out what fractals were needed.
Perhaps these modern day generative programs(sigh, ok ok, "AI") is our generations solution to fractal encoding and the problem with scaling the graduate student algorithm.
"Well we don't exactly know how to solve it but we took 60 Gb of linear algebra and hit it with a stick until it figured it out, no idea how it works but it appears to do the job"
- dreamcompiler 4y agoFractal compression worked automatically (grad student not required) but it typically took hours on the hardware of the late 80s/early 90s. And ISTR it worked well on particular images, but not generally. It's likely it would work much better on modern hardware, but it was heavily IP-encumbered back in the day which caused almost everybody working on image compression algorithms to ignore it. The patents have probably all expired by now so it might be worth another look.
- zimpenfish 4y ago> it was heavily IP-encumbered back in the day It was also horrifically expensive to license from what I remember which meant that almost no-one used it.