TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Remastering Star Trek: Deep Space Nine with Machine Learning

391 pointsby donboxabout 6 years ago

26 comments

CharlesWabout 6 years ago
ST:DS9 was originally shot on film like ST:TNG, but the market performance of the TNG remaster makes it very unlikely that DS9 will receive the same treatment.<p><i>&quot;Essentially, for the HD release of </i>Star Trek<i>, all people had to do was scan each episode. For </i>The Next Generation<i>, they would have to scan all those original pieces of film and then edit together each episode again, themselves. It’s more difficult, more expensive, and much more time-consuming.</i><p><i>&quot;Unfortunately, it wasn’t actually worth it. Sales of the extravagant</i> TNG <i>remaster—original retail price $118 for just season one—failed to reach CBS and Paramount’s expectations. A similar process would have to be done for both</i> DS9 <i>and</i> Voyager<i>—and would actually be even harder.&quot;</i><p><a href="https:&#x2F;&#x2F;io9.gizmodo.com&#x2F;the-detailed-depressing-reason-deep-space-nine-and-voy-1791962332" rel="nofollow">https:&#x2F;&#x2F;io9.gizmodo.com&#x2F;the-detailed-depressing-reason-deep-...</a>
评论 #19456758 未加载
评论 #19455337 未加载
评论 #19454526 未加载
评论 #19456299 未加载
评论 #19455868 未加载
评论 #19456342 未加载
评论 #19457672 未加载
评论 #19456828 未加载
评论 #19455652 未加载
评论 #19455230 未加载
crazygringoabout 6 years ago
Wow, so this is <i>fascinating</i>. Some observations:<p>- Many of the frames look like someone just applied a &quot;sharpen&quot; filter -- there&#x27;s (as expected) no real new information, it&#x27;s just sharper... so it doesn&#x27;t seem like a big deal, like I could do it in Photoshop trivially<p>- But then there are a few spots where new details are truly seamlessly added... the fireball in the spaceship explosion, and forehead wrinkles. Stunning, they&#x27;re absolutely seamless and believable, with detail that is simply not there in the original... that&#x27;s literally magic<p>- But at the same time, characters in the background that are slightly out-of-focus get oversharpened when they&#x27;re supposed to be blurry, like it can&#x27;t tell when moderate blurriness is due to resolution or focus<p>Overall, I&#x27;m pretty shocked that the effect is so seamless across frames -- I totally would have expected this to produce weird discontinuities in time, but I didn&#x27;t see any at all. I mean, this actually seems like it&#x27;s already production-ready to throw into TV&#x27;s or VLC. Which is <i>crazy</i>.<p>I wonder how much of this is &quot;general-purpose&quot;, or to what extent this works on this episode because it&#x27;s trained on this episode? E.g. the neural network is learning from a close-up of the Ferengi face or spaceship, to apply specifically to a later smaller version. Or to what extent this will work well across TV shows, across actors, across genres, without prior training, or with sparse training?
评论 #19461848 未加载
ericsoderstromabout 6 years ago
I don&#x27;t see a huge difference between the two. Mostly the Ferengi forehead creases are sharper I guess, but unclear that the overall effect is much of an improvement. Cool proof of concept though!
评论 #19455026 未加载
评论 #19454557 未加载
评论 #19454121 未加载
评论 #19454153 未加载
评论 #19454128 未加载
评论 #19459863 未加载
jaimebueltaabout 6 years ago
Interestingly, there&#x27;s a project for a documentary about DS9 that&#x27;s close to completion.<p><a href="https:&#x2F;&#x2F;www.indiegogo.com&#x2F;projects&#x2F;what-we-left-behind-star-trek-deep-space-nine-doc" rel="nofollow">https:&#x2F;&#x2F;www.indiegogo.com&#x2F;projects&#x2F;what-we-left-behind-star-...</a><p>It has been delayed due to master the old footage (it&#x27;s one of the last updates)<p>I&#x27;ll leave this work it as a comment on the project.
评论 #19457747 未加载
magdufabout 6 years ago
This is interesting, but it&#x27;s still not the same as a proper remaster. You can&#x27;t get information out of SD video that wasn&#x27;t there before with upscaling, so the algorithm is just interpolating. It&#x27;s too bad that Paramount is so fan-hostile; they could just get the original film footage (assuming it was shot that way; ST:TOS and ST:TNG were, which is why we have excellent remastered versions today), scan it in as raw video, let the fans have at it, and they&#x27;d make the high-def FX for free.
评论 #19454298 未加载
评论 #19454348 未加载
评论 #19455358 未加载
评论 #19456460 未加载
评论 #19454267 未加载
评论 #19454543 未加载
Balgairabout 6 years ago
Pro-tip: You have to change the YouTube setting so that you are watching in 1080 HD, not the 360p that it can default you too.<p>You know, I think there is some real promise here!<p>With some HD remasters, you can start to really see the makeup, the little pores in everyone&#x27;s skin, the smudges and uncleanliness of real film-making. Film-makers and directory choose the lighting and the focus with the end-product in mind. They <i>know</i> that the screen won&#x27;t capture certain things and so they know where they can skimp and save [0]. When you re-master it, you&#x27;re going against the &#x27;vision&#x27; of the directors. Not in a <i>big</i> way at all, it&#x27;s very subtle. But it&#x27;s still there.<p>With ML techniques, you get the &#x27;idea&#x27; that the director was going for, without seeing all the screw-ups that they knew they could get away with. It&#x27;s crisper but the idea and vision are the same [0].<p>Peter Jackson&#x27;s recent &#x27;They Shall No Grow Old&#x27; is another great example of using ML too. In that case it was to preserve the old WW1 footage, bring it&#x27;s frame-rates up to modern 24 fps, 3-D it, and colorize it. The results are literally breath-taking. Personally, I gasped when it finally hits; it&#x27;s that good. Not to geek out too much here, but Jackson is <i>literally changing history</i> with that film. He changed the way we all view old footage, as something all herky-jerky and grainy, to something that is modern and real. Those 16 year old child-soldiers become <i>real people</i> again.<p>Though Jackson&#x27;s work is a lot different than this effort, I think we all know that ML and the movies are here to stay. It&#x27;s relatively cheap to update, takes little time, and be profitable (Remember the Disney Vault gimmick?). How long will it be before Chaplin&#x27;s &#x27;City Lights&#x27; is ML&#x27;d and remastered into 24 fps, 3-D, color, and with sound? Maybe 5 years?<p>Hell, I&#x27;d pay to see the best of old cinema brought back to modern standards like that.<p>[0] I know no film-makers, this is supposition.
评论 #19462486 未加载
kriroabout 6 years ago
What is the licensing situation for remastering like? It would be a pretty interesting competitive edge if a company like Netflix could improve their entire library with some inhouse remastering magic. Or even a bargaining tool for buying new licenses...sure, we&#x27;ll add your stuff to our collection and remaster it &quot;for free&quot;.
评论 #19455526 未加载
conceptionabout 6 years ago
People are using this same technique for old games - <a href="https:&#x2F;&#x2F;venturebeat.com&#x2F;2019&#x2F;01&#x2F;18&#x2F;machine-learning-is-rescuing-old-game-textures-in-zelda-and-final-fantasy&#x2F;" rel="nofollow">https:&#x2F;&#x2F;venturebeat.com&#x2F;2019&#x2F;01&#x2F;18&#x2F;machine-learning-is-rescu...</a>
ocdtrekkieabout 6 years ago
I kinda hope that perhaps, if CBS doesn&#x27;t find value in doing a TNG-level remaster of DS9 and Voyager, that something like this might give them a &quot;good enough&quot; version for a Blu-ray release. At the very least, one could imagine using this technique where applicable to at least, massively cut down the conversion cost in scenarios where re-cutting the scene would be particularly cost prohibitive.
评论 #19454290 未加载
评论 #19454363 未加载
cpetersoabout 6 years ago
&quot;it took me about two days to get everything extracted, upscaled and put it back together in a way that was pleasing. This resulted only in the first five minutes of the episode being done.&quot;<p>Maybe someone at Google can use spare cloud time to remaster the whole series as their 20% time project? This remastering process would be easily parallelizable.
评论 #19460205 未加载
skybrianabout 6 years ago
I see a lot of comments from people getting excited about superficial quality improvements, and I&#x27;m mystified. It&#x27;s nice I guess, but the story and acting are unchanged by this sort of thing, and isn&#x27;t that what matters to Star Trek fans?
评论 #19456191 未加载
Jerry2about 6 years ago
I recently tried to rewatch Babylon 5 but the resolution was terrible and it looked bad on my TV so I gave up pretty quickly. I hope someone does the same treatment to Babylon 5 episodes!
评论 #19455267 未加载
评论 #19455629 未加载
评论 #19455318 未加载
评论 #19459291 未加载
评论 #19457175 未加载
评论 #19460091 未加载
MayeulCabout 6 years ago
As I see more of these machine learning applications, where a neural network is trained on a dataset, then applied to a similar dataset, I wonder how well it would help if the network was trained on part of the new dataset as well, as some kind of &quot;specialization&quot;.<p>One could take a generic &quot;AI Gigapixel&quot; net, and retrain it on some of the newer Star Trek content, for which an upscaled&#x2F;remastered &quot;ground truth&quot; exists. My guess is that it would help a lot with features that are specific to the Star Trek universe.<p>Taking this further, one could take the resulting samples, and &quot;rate&quot; them, before feeding it back in the training engine. This would make some kind of &quot;adversarial&quot; feedback loop, but instead of the GAN, humans are involved in the loop (which could be used to train an adversarial network as well, that said). My hope is that it would converge to much better results in a shorter time, and with less input data.<p>My apologies if those are common concepts in ML. If so, I&#x27;d love to look at some references that could further my understanding on these topics, and get me to use the right terminology.
moultanoabout 6 years ago
This seems like it would really benefit from a model trained specifically to upscale video. There&#x27;s a lot of information you can get from the structure between frames that you can&#x27;t exploit with a still image.
modzuabout 6 years ago
im not discounting ML being able to do this better. kudos to the op. but for fun i tried it using conventional methods (NR+sharpen); similar results but an order of magnitude faster. you be the judge:<p><a href="https:&#x2F;&#x2F;pasteboard.co&#x2F;I6tvYmP.png" rel="nofollow">https:&#x2F;&#x2F;pasteboard.co&#x2F;I6tvYmP.png</a>
AcerbicZeroabout 6 years ago
On the one hand I&#x27;d love to see them remastered and touched up, but on the other hand, I originally watched them on a ~27 inch, old mid 90&#x27;s TV....so the visuals aren&#x27;t really the part which hits my nostalgia nerve.
entropicdrifterabout 6 years ago
Alright, now I want to do this with Babylon 5 since it seems like the HD remaster from WB&#x2F;Amazon petition has flopped. Heck, even just upscaling the SFX shots from 360p to full DVD quality would be a worthwhile project.
nominated1about 6 years ago
I&#x27;m curious as to why the author chose &quot;AI Gigapixel&quot;. What made it a better choice than say NNEDI3 or SuperRes or any other NN upscaler (there’s a bunch of them out there)? Were others even considered?
dbcooperabout 6 years ago
Anyone compared this with MadVR?
bitLabout 6 years ago
Anyone knows how AI Gigapixel differs from DLSS? (outside Topaz Labs ofc). Thanks!
pbhjpbhjabout 6 years ago
Are films&#x2F;programmes now preparing for future re-rendering in any way?
craigmcnamaraabout 6 years ago
Star Trek: Deep Space Nine aka, Cheers in Space!
microDudeabout 6 years ago
Any Friends of DeSoto here?
sigi45about 6 years ago
Holy shit, you did exactly what i wanted to do after i read about the final fantasy results!11 :D
zackkridaabout 6 years ago
You may find the original blog post from the creator here:<p><a href="https:&#x2F;&#x2F;captrobau.blogspot.com&#x2F;2019&#x2F;03&#x2F;remastering-star-trek-deep-space-nine.html" rel="nofollow">https:&#x2F;&#x2F;captrobau.blogspot.com&#x2F;2019&#x2F;03&#x2F;remastering-star-trek...</a>
评论 #19454524 未加载
评论 #19454054 未加载
ChrisArchitectabout 6 years ago
<a href="https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=19406060" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=19406060</a>
评论 #19456264 未加载
评论 #19459115 未加载