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.

Remaking old computer graphics with AI image generation

181 pointsby jerryjerryjerryover 2 years ago

14 comments

Teslazarover 2 years ago
Is there a way to have current AI tools maintain consistency when generating multiple images of a specific creature or object? For example, if there are images of 'Dr. Venom' they need to look similar, or if there are images of the same space ship.
评论 #34213740 未加载
评论 #34213730 未加载
评论 #34218150 未加载
评论 #34213306 未加载
评论 #34231397 未加载
评论 #34217847 未加载
评论 #34213069 未加载
评论 #34214413 未加载
评论 #34213115 未加载
评论 #34216325 未加载
评论 #34215773 未加载
评论 #34217804 未加载
评论 #34214245 未加载
评论 #34213453 未加载
hnbadover 2 years ago
This is a great example of AI image generation being unable to generate &quot;art&quot; and instead just replicating a naive approximation of what it was trained on. There&#x27;s no coherence or consistency between the images and while they all look &quot;shiny&quot;, they also look incredibly dull and generic.<p>It&#x27;s a cool exercise but using this for a real-world project would eliminate any attempt at producing an artistic &quot;voice&quot;. AI image generation excels only at generating stock art and placeholder content.
评论 #34231603 未加载
sirabenover 2 years ago
I was trying this recently with the Sierra Christmas Card from 1986![0] The images that I generated were[1], and I was trying to tweak the model parameters with different denoising and CFG scales. When you get the parameters just right you can preserve the composition of the input image very well while still adding a lot of detail. This isn&#x27;t a completely automatic process though, with Stable Diffusion you have to provide the right prompt otherwise the generation process isn&#x27;t guided correctly, so approach works better for aesthetics and style transfer than regular image super-resolution such as ESRGAN.<p>[0] <a href="https:&#x2F;&#x2F;archive.org&#x2F;details&#x2F;sierra-christmas-card-1986" rel="nofollow">https:&#x2F;&#x2F;archive.org&#x2F;details&#x2F;sierra-christmas-card-1986</a><p>[1] <a href="https:&#x2F;&#x2F;i.imgur.com&#x2F;WxD05gX.jpeg" rel="nofollow">https:&#x2F;&#x2F;i.imgur.com&#x2F;WxD05gX.jpeg</a>
评论 #34214456 未加载
LudwigNagasenaover 2 years ago
That’s just yet another exercise in prompt engineering. I expected something more interesting.
carrolldunhamover 2 years ago
The title excited me - maybe someone succeeded making new art that looks like the old pre-renders, maybe a convincing imitation of the scanline render look. Instead it was a vapid article about tossing pixel art into img-2-img and getting some tenuously related junk.
评论 #34213452 未加载
评论 #34213178 未加载
vyrotekover 2 years ago
This would be super fun to use with &quot;Return to Zork&quot; clips.
shdonover 2 years ago
Is it just me, or was anybody else reminded of the Carmageddon cover art after seeing the image they ended up using for Dr. Venom?
js8over 2 years ago
I have a question. Stable diffusion is based on gradually processing noise into a coherent image, by training a denoiser. Would it be possible to feed low-fidelity image (such as pixel art, or pixelated image) directly into the denoiser step and get a higher-fidelity image that would match the original?
评论 #34212894 未加载
评论 #34212973 未加载
评论 #34213246 未加载
sublinearover 2 years ago
You&#x27;d have to scale the resolution on these waayyy down to not see the usual janky, smudgy, sometimes nightmare-inducing details.<p>I seriously have never understood why what gets published in these blog posts isn&#x27;t just lower res especially since this is precisely about old video game graphics.
评论 #34214405 未加载
评论 #34215573 未加载
评论 #34214265 未加载
xwdvover 2 years ago
It’s amazing that someday we’ll be able to pass in low fidelity pixel art sprite sheets to an AI and get back high definition hand drawn 2D graphics for use in games.
评论 #34212925 未加载
评论 #34213279 未加载
评论 #34212776 未加载
评论 #34212854 未加载
评论 #34212952 未加载
评论 #34212840 未加载
NicoleJOover 2 years ago
What are you going to do about all the copyrighted material?
camtarnover 2 years ago
Honestly, really disappointing, especially since the author forces you to watch the video to see the final image - which looks nothing like the shoulder-spiked, triple-forehead-eye&#x27;d villain of the game. Spoiler, the generated image is just a threatening looking green dude with two different coloured eyes.<p>It&#x27;s a decent writeup on the process of trying to generate specific images using text prompts, I guess, with the conclusion that it&#x27;s really hard, and in some cases basically impossible (hence the lack of the three forehead eyes).
评论 #34216416 未加载
评论 #34216406 未加载
newobjover 2 years ago
Nice, loses all its charm and turns off my imagination.
评论 #34213130 未加载
评论 #34213083 未加载
ggambettaover 2 years ago
At last we have the tech to do a cinematic, photorealistic version of Zero Wing! For great justice!!!
评论 #34216414 未加载