NVIDIA AI Fixing Grainy Pictures

Discussion in '[H]ard|OCP Front Page News' started by Kyle_Bennett, Jul 10, 2018 at 9:09 AM.

  1. Kyle_Bennett

    Kyle_Bennett El Chingón Staff Member

    Messages:
    52,229
    Joined:
    May 18, 1997
    While some of us have some incredibly complex smartphones that take beautiful pictures, there are still those times when lighting causes us to get less than favorable shots. While this tech is not available to us yet, and likely will not be for a while, the folks at Aalto University, MIT, and NVIDIA have been working on a solution to getting those photos touched up without even having to see "clean" ones for comparison. That said, the technology was trained by looking only at corrupted photographs, which is a bit wild when you think about it. While everyone loves a great photo, these techniques are also looking to be used in the medical field, but of course in those instances it has to be exactly right.

    Check out the video.


    “There are several real-world situations where obtaining clean training data is difficult: low-light photography (e.g., astronomical imaging), physically-based rendering, and magnetic resonance imaging,” the team said. “Our proof-of-concept demonstrations point the way to significant potential benefits in these applications by removing the need for potentially strenuous collection of clean data. Of course, there is no free lunch – we cannot learn to pick up features that are not there in the input data – but this applies equally to training with clean targets.”
     
  2. andrewaggb

    andrewaggb Limp Gawd

    Messages:
    293
    Joined:
    Oct 6, 2004
    Some of those are so impressive It's hard to believe. I wonder if the results are typical or cherry picked and how much training is required - would it only work in a cloud scenario or could your phone do it.
     
  3. M76

    M76 [H]ardness Supreme

    Messages:
    6,917
    Joined:
    Jun 12, 2012
    Of course they're cherry picked, what do you think? Promo results are always cherry picked no matter the subject matter. The question is how good the average result is. Which nobody will know until they release the technology. Oh, it is nvidia they won't release it to the public.
     
  4. SomeoneElse

    SomeoneElse [H]ard|Gawd

    Messages:
    1,308
    Joined:
    Jan 16, 2007
    Although most of those are showing impressive results. The overall is impressive but I suspect it will be far from professional quality photos. Too much interpretation is needed to fill in the removed noise and add in the correct color and sharpness.
     
  5. katanaD

    katanaD [H]ard|Gawd

    Messages:
    1,225
    Joined:
    Nov 15, 2016
    WOOHOO !!!

    thanks to nvidia, we will finally see the real bigfoot and all those streetlights !!

    nessie.. here we come
     
  6. xorbe

    xorbe [H]ardness Supreme

    Messages:
    5,842
    Joined:
    Sep 26, 2008
    The MRI one will never happen, they will never allow "best guess AI data back fill" into medical data. Sure the picture looks better and more pleasing, but it's not necessarily accurate after that AI interpretation.
     
    trandoanhung1991 and scojer like this.
  7. xp3nd4bl3

    xp3nd4bl3 2[H]4U

    Messages:
    2,157
    Joined:
    Sep 25, 2004
    I don't think they'd suggest using a corrected image in lieu of the noisy source, but having access to both could certainly be beneficial.
     
  8. Tak Ne

    Tak Ne [H]ard|Gawd

    Messages:
    1,123
    Joined:
    Jan 28, 2008
    Looks to me like the headline should be "Nvidia AI learns to compensate for Nvidia's noise-adding algorithm" :D
     
    clockdogg likes this.
  9. Rahh

    Rahh [H]ard|Gawd

    Messages:
    1,323
    Joined:
    Jan 14, 2005
    I keep seeing Nvidia headlines similar to this which other software already have accomplished. Kudos to them though for creating competition but the fact their software\hardware can accomplish this is nothing new.
     
    clockdogg likes this.
  10. oldmanbal

    oldmanbal [H]ard|Gawd

    Messages:
    1,533
    Joined:
    Aug 27, 2010
    I can see this being abused to remove water marks from pictures taken by the actual author of said content.
     
    Conman likes this.
  11. andrewaggb

    andrewaggb Limp Gawd

    Messages:
    293
    Joined:
    Oct 6, 2004
    I had wondered about that as well.
     
  12. SixFootDuo

    SixFootDuo [H]ardness Supreme

    Messages:
    4,615
    Joined:
    Oct 5, 2004
    Shrug .... I haven't seen grainy photos since 2005
     
  13. brucethemoose

    brucethemoose Limp Gawd

    Messages:
    176
    Joined:
    Mar 3, 2018
    I wish Nvidia would release the source (or at least the compiled versions) of all these demos they do. I'd love to try them out myself.
     
  14. Zarathustra[H]

    Zarathustra[H] Pick your own.....you deserve it.

    Messages:
    24,620
    Joined:
    Oct 29, 2000
    Yeah, that's definitely way better than traditional denoise filters, which typically just amount to a light blur.

    They are almost too good to be true...
     
  15. Zarathustra[H]

    Zarathustra[H] Pick your own.....you deserve it.

    Messages:
    24,620
    Joined:
    Oct 29, 2000
    Try taking pictures in very low light. Concert photography, etc. Granted, the latest very high end DSLR's have remarkably good low noise performance at high ISO, but they are also ridiculously expensive. Even a mid range pro-sumer DSLR, not to mention a cellphone camera is either going to have motion blur or high ISO noise when it gets darker.
     
    Flogger23m likes this.
  16. brucethemoose

    brucethemoose Limp Gawd

    Messages:
    176
    Joined:
    Mar 3, 2018
    Yeah, those demo results are really good, but sharpening after using a good "traditional" denoiser isn't too bad.

    Waifu2X is also a good stand-alone denoiser, but it's neural-net based like this Nvidia demo.

    http://waifu2x.udp.jp
     
  17. Hallucinator

    Hallucinator Limp Gawd

    Messages:
    468
    Joined:
    Nov 1, 2006
    Photoshoppers' wet dream just came true....
     
  18. Retronym

    Retronym Something big is coming.

    Messages:
    23,133
    Joined:
    Mar 5, 2007
  19. raz-0

    raz-0 [H]ardness Supreme

    Messages:
    4,323
    Joined:
    Mar 9, 2003
    Some of the results are also a dumpster fire of oversmoothing. About the only thing I can say about both the good and bad examples is that they appear to largely avoid inducing sharpening artifacts. Which is a plus.
     
  20. lostin3d

    lostin3d [H]ard|Gawd

    Messages:
    1,137
    Joined:
    Oct 13, 2016
    Pretty cool but I wonder has this AI contacted Ted Turner to further his touch up projects?

     
  21. Uvaman2

    Uvaman2 2[H]4U

    Messages:
    2,272
    Joined:
    Jan 4, 2016
    The corrupted example was pretty impressive. So those with no additional data huh? Nice.
     
  22. Ski

    Ski Gawd

    Messages:
    819
    Joined:
    Jun 21, 2008
    You guys should check out V-Ray's latest engine update with Nvidia's denoiser incorporated inside it. It's very excited for for the cgi and animation industry.
     
  23. Decibel

    Decibel 2[H]4U

    Messages:
    3,741
    Joined:
    Nov 10, 2000
  24. Luke Wells

    Luke Wells n00bie

    Messages:
    23
    Joined:
    Jan 24, 2018
    CSI Miami is becoming a reality!

    "Computer..... Enhance resolution."

    "Well, I guess we ... resolved this case.... " [puts on sunglasses]

    [Cue the music]
     
  25. Trepidati0n

    Trepidati0n [H]ardForum Junkie

    Messages:
    12,106
    Joined:
    Oct 26, 2004
    Then you haven't taken a photo with a smart phone at night ever.
     
  26. jpm100

    jpm100 [H]ardness Supreme

    Messages:
    6,959
    Joined:
    Oct 31, 2004
    All you nay sayers about Sci-Fi "Zoom & Enhance", take that!
     
  27. lostin3d

    lostin3d [H]ard|Gawd

    Messages:
    1,137
    Joined:
    Oct 13, 2016
    This could really help Decker for the next BR movie.
     
  28. oldmanbal

    oldmanbal [H]ard|Gawd

    Messages:
    1,533
    Joined:
    Aug 27, 2010
    anything from an iphone 6 and up have excellent photos. The 5 still takes cheese a lot at distance.
     
Tags: