Wavelets (1994) [pdf] • Hacker News Board

Re: Wavelets (1994) [pdf]

#31

Post by rasz » Thu, Aug 15, 2019, 7:40 AM UTC

patentatt wrote: post #3

Wavelets were supposed to be a big thing like 15-20 years ago ... what happened? Do any mainstream modern codecs use wavelets? I’m not aware of any Edit: looks like the Dirac video codec is based on wavelets. Good to know all of that research wasn’t for nothing!

DjVu ebook format uses wavelet compression. From my experience its painfully slow.

Re: Wavelets (1994) [pdf]

#32

Post by pavlov » Thu, Aug 15, 2019, 8:46 AM UTC

goalieca wrote: post #14

patentatt wrote: post #3

Wavelets were supposed to be a big thing like 15-20 years ago ... what happened? Do any mainstream modern codecs use wavelets? I’m not aware of any Edit: looks like the Dirac video codec is based on wavelets. Good to know all of that research wasn’t for nothing!

Wavelets are the basis of modern video codecs. People also use them to watermark videos. I never found much advanced application in image processing which was my field of research.

There's no wavelet-based video codec in widespread use (unless one counts DCP, a file format used for physical deliveries of movies to theaters, where the picture track is encoded as JPEG2000 frames).

Re: Wavelets (1994) [pdf]

#33

Post by vanderZwan » Thu, Aug 15, 2019, 8:48 AM UTC

patentatt wrote: post #3

Wavelets were supposed to be a big thing like 15-20 years ago ... what happened? Do any mainstream modern codecs use wavelets? I’m not aware of any Edit: looks like the Dirac video codec is based on wavelets. Good to know all of that research wasn’t for nothing!

Not a codec, but still image-related: one of the most powerful modules in Darktable is the equalizer, which is based on wavelets[0][1].

[0] http://www.darktable.org/2011/11/darktable-and-research/

[1] https://www.roberthutton.net/archives/203

Re: Wavelets (1994) [pdf]

#34

Post by markus92 » Thu, Aug 15, 2019, 8:55 AM UTC

userbinator wrote: post #27

zyxzevn wrote: post #18

Earlier quoted context omitted.

Blocked by Patent-bullshit

I believe either the patents have expired or the basic functionality was always royalty-free; but what really killed JPEG2000 was the immense complexity and increased processing power required. It still has applications in some niche areas, and probably most people's contact with it will be in the form of images embedded in PDFs (many of archive.org's ebooks use JPEG2000), where the noticeably slower rendering of pag…

Also digital cinema is pretty much all JPEG2000 (every frame is individually compressed).

Re: Wavelets (1994) [pdf]

#35

Post by mrcactu5 » Thu, Aug 15, 2019, 9:44 AM UTC

how does this compare to Fast Fourier transform ?

https://en.wikipedia.org/wiki/Fast_Fourier_transform

Or the Discrete Cosine Transform ?

https://en.wikipedia.org/wiki/Discrete_cosine_transform

Or the Gabor Transform ?

https://en.wikipedia.org/wiki/Gabor_transform

Re: Wavelets (1994) [pdf]

#36

Post by mirsadm » Thu, Aug 15, 2019, 9:45 AM UTC

mjfl wrote: post #13

Anyone know good sources on shift invariant wavelets?

The dual tree complex wavelet transform is approximately shift invariant https://en.wikipedia.org/wiki/Complex_wavelet_transform#Dual...

Re: Wavelets (1994) [pdf]

#37

Post by woliveirajr » Thu, Aug 15, 2019, 10:26 AM UTC

I love how you can decompose one image in wavelets and edit just some layers of it (in GIMP), and those retouches end up as subtle improvements of the image.

Re: Wavelets (1994) [pdf]

#38

Post by conjectures » Thu, Aug 15, 2019, 10:45 AM UTC

mrcactu5 wrote: post #35

how does this compare to Fast Fourier transform ? https://en.wikipedia.org/wiki/Fast_Fourier_transform Or the Discrete Cosine Transform ? https://en.wikipedia.org/wiki/Discrete_cosine_transform Or the Gabor Transform ? https://en.wikipedia.org/wiki/Gabor_transform

In what sense?

Re: Wavelets (1994) [pdf]

#39

Post by jhbadger » Thu, Aug 15, 2019, 11:03 AM UTC

patentatt wrote: post #3

Wavelets were supposed to be a big thing like 15-20 years ago ... what happened? Do any mainstream modern codecs use wavelets? I’m not aware of any Edit: looks like the Dirac video codec is based on wavelets. Good to know all of that research wasn’t for nothing!

There was even a whole pop-sci book -- Hubbard's "The World According to Wavelets" (1996) that was a bit like Mandelbrot's "The Fractal Geometry of Nature" in that it predicted an amazing future for the technique and claimed that science would be revolutionized by it. But it is nice to see that wavelets, like fractals, do have their uses even if the revolution never came.

Re: Wavelets (1994) [pdf]

#40

Post by zrav » Thu, Aug 15, 2019, 2:34 PM UTC

patentatt wrote: post #3

Wavelets were supposed to be a big thing like 15-20 years ago ... what happened? Do any mainstream modern codecs use wavelets? I’m not aware of any Edit: looks like the Dirac video codec is based on wavelets. Good to know all of that research wasn’t for nothing!

According to the x264 developers (can't find the reference anymore, sorry), wavelet based lossy compression is good for optimizing for PSNR, not for perceptual image/audio quality, being prone to loss of detail. x264 beats Dirac and modern image compression formats like HEIF (based on HEVC) easily beat JPEG2000. Modern codecs usually use transforms derived from the classic DCT and DST.