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.

AI generation of fake celeb images

81 pointsby nsouthabout 7 years ago

13 comments

symphomabout 7 years ago
Amazing. And now one only needs a random article generator, which describes marriages, breakups, mishaps, accidents, and plain random appearances of these celebrities. Tada, random fake celebrity news, which is probably even better at wasting people's time than the real thing. This is going to happen really soon, because text generation is much easier than what the Nvidia guys did.
评论 #16895986 未加载
SteveJSabout 7 years ago
The thumbstamp version of creating fake celeb faces was one of the projects for the Deep Learning Udacity nanodegree. (These get changed over time, so it may be different now.)<p>The dataset of 200,000 celeb photos with the face nicely centered at a known location is a nontrivial part of making the exercise feasible.<p>I trained on windows with a 6Gb 1060, and went off script from the DCGAN paper, by using upscaling rather than transpose convolutions. Once all the fiddling details are set correctly, the results are quite amazing. It didn’t even require a complete single pass over that dataset.
prionsabout 7 years ago
Really interesting article seeing how I&#x27;m finishing up my own DCGAN project!<p>Generative models like GANs are fascinating, but very temperamental to train. Some of the findings in this paper mirror my own observations - increasing the complexity of the GAN adds a lot of instability. My solution was to keep things as simple as possible. I spent a lot of effort trying to increase the size of the network to get better results, but in the end my smallest implementation worked the best.<p>This bit is interesting: &quot;Without progressive growing, all layers of the generator and discriminator are tasked with simultaneously finding succinct intermediate representations for both the large-scale variation and the small-scale detail. With progressive growing, however, the existing low-resolution layers are likely to have already converged early on, so the networks are only tasked with refining the representations by increasingly smaller-scale effects as new layers are introduce&quot;
VikingCoderabout 7 years ago
I think this would be fun to generate character portraits for role-playing games.
flossballabout 7 years ago
Coming soon as the next CEO and CTO of your favorite amazing ICO.
评论 #16896643 未加载
enrmarcabout 7 years ago
I read the paper, but I did not understand a thing. What is the path to follow (for example, books or papers to read) in order to at least understand what the paper is talking about? My background is in computer science.
评论 #16896474 未加载
评论 #16896311 未加载
laingaabout 7 years ago
From the demonstration video [0] in the link, very interesting, but also very noticeable where the GAN attempts glasses or other &quot;items&quot; (see around 0:58 and 2:35, respectively). Has any research been done into GANs which aim for realistic artificial objects, instead of faces?<p>[0] <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=G06dEcZ-QTg&amp;feature=youtu.be" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=G06dEcZ-QTg&amp;feature=youtu.be</a>
评论 #16897490 未加载
igraviousabout 7 years ago
&quot;We recommend NVIDIA DGX-1 with 8 Tesla V100 GPUs.&quot;<p><a href="https:&#x2F;&#x2F;www.nvidia.com&#x2F;en-us&#x2F;data-center&#x2F;dgx-1&#x2F;#order-now-dgx-1" rel="nofollow">https:&#x2F;&#x2F;www.nvidia.com&#x2F;en-us&#x2F;data-center&#x2F;dgx-1&#x2F;#order-now-dg...</a><p>The NVIDIA DGX is available for purchase in select countries and is priced at:<p><pre><code> DGX with P100 at $129,000* DGX with V100 at $149,000* DGX support plan is required and must be purchased separately.</code></pre>
评论 #16897848 未加载
nickcwabout 7 years ago
It is interesting really studying the two images on the github page. My first thought was - wow, amazing! But having looked at them for a bit longer they&#x27;ve fallen straight into the uncanny valley for me.<p>The female&#x27;s left and right eyes are different shapes, as are her eyebrows. And the male&#x27;s ears appear to be in different places on the left and right side of his head. His eyes are creepily different too.
评论 #16897943 未加载
pieterrabout 7 years ago
The accompanying video reminds me of the Godley and Creme video ‘Cry’ from 1985.<p>Nvidia article: <a href="https:&#x2F;&#x2F;youtu.be&#x2F;G06dEcZ-QTg" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;G06dEcZ-QTg</a><p>Cry: <a href="https:&#x2F;&#x2F;youtu.be&#x2F;KxtPRF6NG7I" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;KxtPRF6NG7I</a>
评论 #16896475 未加载
phjesusthatguy3about 7 years ago
The fakes look really sweaty&#x2F;shiny. Like they&#x27;re fan photos posted to wikipedia.
ferrosabout 7 years ago
Random? The girl looks exactly like Holly Valance.
评论 #16899108 未加载
评论 #16898614 未加载
ssalkaabout 7 years ago
Combine this with deep fakes and you&#x27;ve got a very dangerous pairing of new technologies