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.

DeepPrivacy: A Generative Adversarial Network for Face Anonymization

135 pointsby baylearnover 5 years ago

10 comments

jsilenceover 5 years ago
Wow, that is really close to the animation of the anonymizer hood in &#x27;a scanner darkly&#x27; (<a href="https:&#x2F;&#x2F;www.imdb.com&#x2F;title&#x2F;tt0405296&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.imdb.com&#x2F;title&#x2F;tt0405296&#x2F;</a>).
评论 #20976892 未加载
coolspotover 5 years ago
Would be useful for amateur adult video sharing. Where face privacy is important, but blur kills the joy.
评论 #20978436 未加载
评论 #20976082 未加载
pndyover 5 years ago
That example gif at the top - the way faces blend between reminds me of camouflage effect from &quot;A Scanner Darkly&quot; film
评论 #20976982 未加载
false-mirrorover 5 years ago
Really cool project! I wonder how long until this sort of software is used in place of blurred-faces.<p>It would be really nice to shoot a public space and instead of worrying about release forms, just anonymize the footage.
kzrdudeover 5 years ago
I didn&#x27;t know what GANs were or how they worked, but this video from yesterday explained it really well. So I&#x27;ll recommend it for anyone else curious!<p>The clear explanation of what a GAN is from the first 8-10 minutes of the video. And there are quite a few interesting examples in there as well.<p><a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=dCKbRCUyop8" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=dCKbRCUyop8</a>
thomas_baiover 5 years ago
Great to see more research being done on the topic! :) At Brighter AI, we work on this technology since almost 2 years, offering a natural anonymization solution for faces as well as license plates (<a href="https:&#x2F;&#x2F;brighter.ai&#x2F;video" rel="nofollow">https:&#x2F;&#x2F;brighter.ai&#x2F;video</a>). Preventing facial recognition when collecting data in public is very important and to be fair, also other companies pursue this now.<p>Best regards from Berlin, Thomas (thomas@brighter.ai)
a785236over 5 years ago
I wish the authors wouldn&#x27;t oversell the privacy claim:<p>&gt; Github: &quot;The DeepPrivacy GAN never sees any privacy sensitive information, ensuring a fully anonymized image.&quot;<p>&gt; Abstract: &quot;We ensure total anonymization of all faces in an image by generating images exclusively on privacy-safe information.&quot;<p>&gt; Paper: &quot;We propose a novel generator architecture to anonymize faces, which ensures 100% removal of privacy-sensitive information in the original face.&quot;<p>Changing a face anonymizes an image the same way that removing a name anonymizes a dataset -- poorly. This is cool, but it&#x27;s not anonymization.
评论 #20978485 未加载
baylearnover 5 years ago
link to the arxiv paper: <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1909.04538" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1909.04538</a>
optician_owlover 5 years ago
Looks like ears are unchanged but it&#x27;s a strong identifier.
dvhover 5 years ago
DeepGoldstein