[HN Gopher] AI-Powered Nvidia RTX Video HDR Transforms Standard ...
       ___________________________________________________________________
        
       AI-Powered Nvidia RTX Video HDR Transforms Standard Video into HDR
       Video
        
       Author : Audiophilip
       Score  : 85 points
       Date   : 2024-01-24 16:04 UTC (6 hours ago)
        
 (HTM) web link (blogs.nvidia.com)
 (TXT) w3m dump (blogs.nvidia.com)
        
       | zamadatix wrote:
       | A bit of a let down that the video demoing SDR->HDR conversion is
       | itself only published in SDR. Makes as much sense as demoing a
       | colorization tool in a grayscale video!
        
         | rado wrote:
         | Ridiculous. Like when James Cameron promoted Avatar HDR with an
         | SDR YouTube video, while YT is perfectly capable of HDR
         | playback.
        
           | babypuncher wrote:
           | At least as of a couple of years ago, HDR support on YouTube
           | has been pretty bad[1]. I know they've been working to
           | improve things since, but I kind of don't blame people for
           | walking away from that mess.
           | 
           | 1. https://www.youtube.com/watch?v=DwDWQyBF9II
        
             | rado wrote:
             | Thanks, will check out LTT's gripes, but I've been watching
             | the following HDR channels forever and they look great:
             | 
             | https://www.youtube.com/@SeoulWalker
             | https://www.youtube.com/@Rambalac
             | https://www.youtube.com/@Relaxing.Scenes.Driving
        
               | m4rtink wrote:
               | Can't recommend Rambalac enough - I have pretty much re-
               | traced his steps multiple times during our Japan trip a
               | couple times & it really helped with orientation. :)
               | 
               | Also some of the walks are really interesting & really
               | gives you the context of various places in Japan. :)
        
               | eurekin wrote:
               | The real issue is it's either HDR or good SDR, but not
               | both at the same time
        
         | sharperguy wrote:
         | At this point, with any new model I think it makes sense to
         | wait until you can run the model on your own input before
         | making any assumptions based on cherry picked examples.
        
         | kelseyfrog wrote:
         | I guess. There's a lot of details we don't know that would
         | change the calculus on this.
         | 
         | To use a analogous workflow, it could be like saying, "It's
         | pointless to shoot video in 10-bit log if it's going to be
         | displayed on Rec.709 at 8-bits." It completely leaves out
         | available transforms and manipulations in HDR that do have a
         | noticeable impact even when SDR is the target.
         | 
         | Again, we can't know if it's important given the information
         | that's available, but we can't know if it's pointless either.
        
         | CamperBob2 wrote:
         | YouTube tends to post a downscaled SD version first, then they
         | encode and post the higher-res versions when they get around to
         | it. This can take days in some cases. Meanwhile the creator
         | catches the flak...
        
           | unsane wrote:
           | Creators don't publish videos until the high-res versions are
           | done processing.
        
           | zamadatix wrote:
           | You don't need high res for HDR on YouTube (144p HDR is a
           | thing there oddly enough) and the 4k version had already
           | processed when I posted that comment (with no change since in
           | HDR availability). Usually media announcements/large channels
           | pre-upload the video so it's ready when they want it to
           | actually publish to avoid that kind of issue though.
        
         | kevingadd wrote:
         | HDR video playback in the browser is pretty unreliable unless
         | you're on a Mac.
        
           | ffsm8 wrote:
           | It's also pretty unreliable on Mac too...
           | 
           | It's more reliable then on linux though, and windows has been
           | doing "auto HDR" for videos for years, so kinda hard to tell
           | when something is HDR or not there.
        
           | devwastaken wrote:
           | HDR through YouTube appears to work fine even on my non HDR
           | certified HDR monitor.
        
           | zamadatix wrote:
           | In what way? I've been doing it without issue on PC longer
           | than I've even owned a Mac.
        
         | mysteria wrote:
         | If they were serious about showing this tech off they should've
         | provided a video file download. Also indicate that it's a HDR
         | file and should only be viewed on a HDR display. Youtube is
         | just making this look bad as people won't see a difference.
        
       | kwanbix wrote:
       | The HDR transformation was really impresive. The Upscale not so
       | much. At least not in my monitor.
        
       | zokier wrote:
       | > Using the power of Tensor Cores on GeForce RTX GPUs, RTX Video
       | HDR allows gamers and creators to maximize their HDR panel's
       | ability to display vivid, dynamic colors, _preserving intricate
       | details_ that may be inadvertently _lost due to video
       | compression_.
       | 
       | There is so much marketing BS in one small paragraph. For
       | starters, generating(/hallucinating) data is imho the opposite of
       | _preserving_ anything. Then HDR is less associated with
       | "intricate details" and more to do with color reproduction.
       | Finally, video compression is the one thing that usually does
       | _not_ have problems with HDR, even the now venerable x264 can
       | handle HDR content, generally it 's almost everything else that
       | struggles.
       | 
       | Of course in a true marketing tradition, none of the things are
       | also strictly false. I'm sure there are many ways to weasel the
       | claims.
        
         | nwellnhof wrote:
         | They claim to preserve color detail that was lost due to
         | compression of the dynamic range. What's wrong with that?
        
           | dbspin wrote:
           | Not the OP, but you have to understand that 'compression of
           | the dynamic range' is an artistic tool. Literally choosing
           | the lighting ratio of an image is how you build out lighting
           | for a scene. With AI overwriting these choices, you're
           | looking at something more akin to colorization than
           | upscaling.
        
             | Grafikenjo wrote:
             | I think you need to understand that it is not always a
             | 'artistic tool' but a money or knowledge limitation.
        
               | dbspin wrote:
               | I'm a filmmaker... I don't know a single DOP or director
               | who wishes they could work in HD but is limited by
               | finances or knowledge. Again, shaping light is the
               | essence of cinematography. Modern DSLRs far surpass the
               | dynamic range (although not the effective resolution) of
               | 35mm film. And yet the image they produce isn't
               | comparable. When it comes to image quality, bit depth is
               | enormously more important than dynamic range. When it
               | comes to creating an artistic image, dynamic range hasn't
               | been a limit for many decades.
        
             | mirsadm wrote:
             | Not really, half the battle with SDR video is tonemapping a
             | high dynamic range to fit into SDR. That process is not
             | artistic, it's a process on trying not to make it look bad.
        
           | jl6 wrote:
           | You can't preserve something that was lost (but perhaps you
           | can recreate a substitute).
        
       | skottenborg wrote:
       | I could see a future where this works really well. It doesn't
       | seem to be the case right now though.
       | 
       | The "super resolution" showcased in the video seemed almost
       | identical to adjusting the "sharpness" in any basic photo editing
       | software. That is to say, perceived sharpness goes up, but actual
       | conveyed details stays identical.
        
         | brucethemoose2 wrote:
         | Note that YouTube is really bad for these demos due to the re-
         | compression, even in zoomed in stills.
        
         | thefourthchime wrote:
         | Allegedly the new one plus phone does this trick in real time
         | as well as up sampling and interframe motion interpretation.
         | Mrwhostheboss seems impressed, but I don't really trust his yet
         | judgment on these things.
         | 
         | https://www.youtube.com/watch?v=J9-9fP_pcEc&t=1107s
        
           | nomel wrote:
           | The iPhone has also done this, for a few years now. It was,
           | surprisingly, a one sentence mention in the keynote/release
           | notes.
        
         | moondev wrote:
         | Whatever special sauce the Nvidia shield uses is honestly
         | incredible. Real time upscaling of any stream, and not just
         | optimized for low res source, its like a force multiplier on
         | content that is already HD. Supposedly the windows drivers do
         | it as well but the effect seems less noticeable to me in my
         | tests
        
       | varispeed wrote:
       | That seems like a gimmick and I actually prefer SDR video that is
       | not upscaled. There is something ugly about those AI treated
       | videos. They look fake.
        
         | deergomoo wrote:
         | They _are_ fake. Ultimately it's not recovering lost detail,
         | it's making shit up
        
           | nomel wrote:
           | I don't think making things up is the problem, it's if it's
           | believable. If it's indistinguishable to a viewer, then who
           | cares. I never would have thought the HDR of the clouds was
           | "made up".
        
             | SirMaster wrote:
             | Maybe I'm odd, but a big part of art to me is seeing things
             | how the creator intended it to be seen.
             | 
             | So I calibrate all my media consumption displays etc. I
             | could never see myself using some automated SDR -> HDR
             | conversion like this.
             | 
             | Even if it looks natural, it doesn't look like it was
             | supposed to, and I want to see it how it was supposed to
             | look.
        
               | nomel wrote:
               | I would use it on every single video I've ever made
               | myself, because intent had nothing to do with how my
               | videos look. They were made with then best camera I had
               | available, and HDR has only been available relatively
               | recently.
               | 
               | This is a tool that I want to use. If you don't want to
               | use it, then nobody is making you.
        
             | UberFly wrote:
             | "I don't think making things up is the problem, it's if
             | it's believable"
             | 
             | Depends of course if it's being passed off as reality.
             | Slippery slope and all.
        
           | zeusk wrote:
           | These remind me of the Samsung debacle about recognizing moon
           | and emplacing a high quality texture of it into the image
           | shot by camera.
        
           | 4d4m wrote:
           | Exactly. This is akin to upscaling or frame rate
           | interpolation. No consumers want this, they turn it off in
           | settings.
        
       | aantix wrote:
       | I'm curious - what's the best open-source video upscaling library
       | out there?
       | 
       | I looked back about a year ago, and it didn't seem like there
       | were any good open-source solutions.
        
         | cf100clunk wrote:
         | An HN search of ''Deep Space Nine'' and ''Topaz'' will show
         | some great discussions here covering the dearth of such
         | upscaling solutions, as well as some huge efforts before
         | commonplace AI.
        
         | adzm wrote:
         | Topaz is light years ahead of any open source solution
         | unfortunately.
        
         | two_in_one wrote:
         | It depends on what do you mean by 'open-source', along with
         | training materials and full setup? That will be hard to find.
         | Upscaling was popular like 10 years back. That's why there is
         | no much interest today. Training in old style isn't that hard.
         | But artifacts are popping up in all videos I've seen.
        
         | justinclift wrote:
         | It's not exactly what you're after, as it's anime specific and
         | you need to process the video yourself (eg disassemble to
         | frames, run the upscaler, then assemble back to a movie file),
         | but Real-ESRGAN is very good for cleaning up old, low
         | resolution anime:
         | 
         | https://github.com/xinntao/Real-ESRGAN/
        
       | manmtstream wrote:
       | Now it will be absolutely impossible to accurately convey the
       | artistic intent, when there's no way to know how it will look on
       | consumer devices.
        
         | cmcconomy wrote:
         | I think we lost that battle with motion interpolation on
         | consumer TVs
        
           | WithinReason wrote:
           | Already happened with brightness and contrast controls
        
             | cm2187 wrote:
             | I can't get any two computer monitors that are not the same
             | model to give me the same color.
        
         | luma wrote:
         | Consumer devices have never been known for color accuracy and
         | goes back a very long ways. The running joke in broadcast was
         | that NTSC stood for "Never Twice the Same Color".
        
       | aaroninsf wrote:
       | The work I am interested in this broader domain is conversion
       | (say, via some NeRF) of existing standard video into spatial
       | video e.g. MV-HEVC for immersive experience on the Vision Pro
       | etc.
        
       | rixrax wrote:
       | I recently had some old super8 films shot by my parents scanned
       | into 1080p resolution in ProresHQ. Because of the poor optics of
       | the original camera, imperfect focus when shooting, poor
       | lightning conditions, and general deterioration of the film
       | stock, most of the footage won't get anywhere near what 1080p
       | could deliver.
       | 
       | What I'd like to try at some point is to let some AI/ML model
       | process the frames, and instead of necessarily scaling it up to
       | 4k etc., 'just' add (aka magic) missing detail into 1080p version
       | and generally unblur it.
       | 
       | Is there anything out there, even in research phase that can take
       | existing video stock, and then hallucinate into it detail that
       | never was there to begin with? What NVidia is demoing here seem
       | like steps to that direction...
       | 
       | I did test out Topaz Video and DaVinci's built-in super
       | resolution feature, both of which gave me a 4k video with some
       | changes to the original. But not the magic I am after.
        
         | baq wrote:
         | there is AI tech to do this already. it has a slight problem,
         | though: it _adds detail_ to faces (this is marketing speak for
         | _completely changes how people look_ ).
        
         | UberFly wrote:
         | Something like this will always change the original as it's
         | guessing what should be there as it up scales. Only time will
         | improve the guessing.
        
         | actionfromafar wrote:
         | An interesting thing about Super8: the resolution is generally
         | very poor, but it can have quite the dynamic range. Also, with
         | film in general (and video, but it's easier with film because
         | you have global shutter) you can compensate motion blur and get
         | more detail out which isn't visible when you look at the film
         | frame by frame. And none of this needs AI.
         | 
         | Regarding hallucination, I agree with the sibling comment, the
         | problem is that faces change. And with video, I'm not even sure
         | the same person would have the same face in various parts of
         | the video...
        
       | bendergarcia wrote:
       | I think they should rephrase. It makes SDR appear HDR. It's just
       | making up information no? It's not actually making it HDR just it
       | appears to be HDR?
        
         | Alghranokk wrote:
         | Making up information? The same can be said for most commonly
         | used modern compressed video formats. Just low bitrate streams
         | of data that gets interpolated and predicted into producing
         | what looks like high resolution video. AV1 even has entire
         | systems for synthesizing film grain.
         | 
         | The way i see it, if the ai generated HDR looks good, why not?
         | It wouldn't be more fake or made up than the rest of the video.
        
       | renewiltord wrote:
       | This stuff is sick. If we had a real-time upscaler on a zoom
       | telescope it would be a fantastic tool while traveling. I'd get a
       | kick out of that.
        
         | genman wrote:
         | And what would fake detail in the real world give to you?
        
           | arcticfox wrote:
           | it doesn't have to be "fake" detail, an AI can use multiple
           | frames to gather much more information than is available in a
           | single frame and composite them into a much more detailed
           | image
        
           | renewiltord wrote:
           | I can pretty easily distinguish useful LLM output from non-
           | useful LLM output though others on this website seem to have
           | lots of trouble. I think I can pretty easily do the same for
           | things in the visual field. To be honest, part of why I'm
           | successful is that I can draw use out of imperfect tools.
           | 
           | e.g. we know UDP has no delivery guarantees, but I can build
           | a pretty good price feed from that, and fast enough that it
           | makes money. Many HN users probably couldn't do that because
           | "UDP has no delivery guarantees! How can you build a book?!
           | It would be fake!". Yeah, well, I can. It's part of why our
           | prop shop makes money.
           | 
           | e.g. No single photograph in a stacked astrophotograph has
           | the info, but you can get a pretty useful image out of it.
        
         | xcv123 wrote:
         | Traveling to a real destination so that you can look at fake AI
         | generated crap on a screen instead of the actual surroundings.
        
       | DrNosferatu wrote:
       | Speaking of which, Nvidia has built-in live AI upscaling on the
       | Shield TV android box.
       | 
       | - Is there any stand-alone live AI upscaling / 'enhance'
       | alternative for android or any other platform?
        
         | lagadu wrote:
         | The Shield is kind of an extreme outlier in today's
         | environment. A device from 2015 that 9 years later is still one
         | of the top tier choices in its (consumer) market is almost
         | unheard of.
         | 
         | In fact it's reportedly the currently supported Android device
         | out there with the longest support[0], it's crazy that mine
         | still gets updates.
         | 
         | [0]https://www.androidcentral.com/android-longest-support-
         | life-...
        
           | moondev wrote:
           | It really is awesome. I also enjoy the UI that allows you to
           | side by side compare a stream and the difference is insane.
           | 
           | I have been meaning to see how well it handles streaming a
           | desktop via moonlight to the shield to real time upscale a
           | second monitor's content. I assume it's trained for video
           | footage and not static UI components. The RTX windows drivers
           | don't seem to upscale as well as the shield.
        
       | fsiefken wrote:
       | I wonder if AI can be used to extrapolate 4:3 to 16:9 format or
       | to create stereoscopic video (for use in VR or 3D TV's)
        
         | LeoPanthera wrote:
         | During the brief moment that 3DTV was popular, almost all 3DTVs
         | had a mode that could "convert" 2D to 3D, based on movement in
         | the scene and other pre-learned cues. "Things that look like
         | people should be in front of things that look like scenery",
         | and so on.
         | 
         | I miss 3D. I loved it, and I was sad that it didn't catch on.
         | It enjoyed a longer life in Europe, where 3D blu-rays were
         | produced for a few more years after they stopped selling them
         | in the US, and I imported and enjoyed several.
         | 
         | Maybe Apple's VR headset will be a 3D renaissance.
        
           | jedberg wrote:
           | The main reason at home 3D failed is because most people
           | don't watch at home like they do at a theater.
           | 
           | At a theater you sit down knowing that you can't get up and
           | leave until it's over. At home you are doing other things:
           | eating, folding laundry, going to the bathroom, taking phone
           | calls, answering the door, and so on. It's not conducive to
           | wearing glasses.
           | 
           | Vision will have the same problem (as does any at home
           | headset). I don't think it will lead to a 3D renaissance, at
           | least not for a long time, until it becomes acceptable (and
           | feasible) to walk around with it on all the time.
           | 
           | Otherwise we need to wait for holographic projectors that can
           | make a 3D image without having to wear glasses that make it
           | hard or impossible to look at other 3D objects.
        
             | pawelduda wrote:
             | I think that would be a problem with VR headsets, not these
             | 3D glasses you could put on or take off in seconds?
        
               | jedberg wrote:
               | My parents had a 3D TV. It was a huge problem. You
               | probably don't realize how often you look away from your
               | TV when you're watching your TV.
        
               | pawelduda wrote:
               | While watching a movie, look away - maybe. Get up and
               | walk around and do chores - we always pause if needed. I
               | think it's a matter of establishing that to watch a
               | movie, one needs to set aside time and commit to focus
               | just on it, but then it becomes yet another barrier.
               | 
               | Different story for TV shows which often are background
               | though.
        
               | jedberg wrote:
               | Yeah that's my point -- most people don't watch movies at
               | home that way. They are just background distractions,
               | even movies.
        
               | naasking wrote:
               | > My parents had a 3D TV. It was a huge problem.
               | 
               | A huge problem if you're not actually watching the movie,
               | sure. But if you're doing something else, don't use the
               | 3D mode.
        
               | jedberg wrote:
               | That's my point -- most people don't actually watch the
               | movie at home. That's why 3D TV failed.
        
               | naasking wrote:
               | I don't think that's why. TVs without 3D are just
               | cheaper, the early 3D tech just wasn't very good and took
               | awhile to mature thus souring the market, 3D content was
               | more expensive (or an extra expense, eg. buy the 3D and
               | non-3D versions of a movie) and so people just went for
               | the cheaper options overall. I've had an active 3D TV for
               | 10+ years, and the 3D has not itself been a problem when
               | I've watched with others.
               | 
               | The only time it's a problem is if someone currently
               | experience a migraine is trying to watch, then they can
               | get serious vertigo, but that's an issue caused by the
               | migraine itself (visual auras and vertigo generally).
        
               | jedberg wrote:
               | All of those reasons certainly contributed, but the
               | reality is that most people don't watch movies at home
               | the way they watch in a theater, where they dedicate 2+
               | hours to the experience with no distractions.
               | 
               | I do that, but I have to wait for everyone to go to bed
               | first and then turn off my cell phone. Most people aren't
               | willing to do that.
        
               | pawelduda wrote:
               | A lot of people wrote it off as unnecessary gimmick, I'd
               | add that to list of reasons. VR 3D blows it out of water
               | but then requires more effort to use.
        
         | naasking wrote:
         | Possibly to some degree. They're doing crazy things with NeRF.
        
       | 4d4m wrote:
       | Feels like a misnomer, its really "HDR style" video. The source
       | material does not have the dynamic range embedded, this is an
       | effect filter.
        
       | cm2187 wrote:
       | So now we need to stop making fun of cops pressing the "enhance"
       | button in films...
        
       ___________________________________________________________________
       (page generated 2024-01-24 23:01 UTC)