<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Unsupervised Image-to-Image Translation Networks · HackerPapers discussions</title><link>https://hackerpaper.kaipfister.de/p/1703.00848/</link><description>Collected HN comments and submissions linking this paper.</description><language>en</language><lastBuildDate>Mon, 02 Nov 2026 11:16:22 GMT</lastBuildDate><atom:link href="https://hackerpaper.kaipfister.de/p/1703.00848/feed.xml" rel="self" type="application/rss+xml" /><item><title>Unsupervised Image-to-Image Translation Networks — mentioned by claytonius</title><link>https://news.ycombinator.com/item?id=21804173</link><guid isPermaLink="false">urn:hackerpapers:mention:1703.00848:21804173</guid><pubDate>Mon, 16 Dec 2019 16:24:23 GMT</pubDate><description>&lt;p&gt;I think you&amp;#x27;re right that most DeepFakes use Variational AutoEncoders (As described by the article) where the latent space is jointly distributed between the two domains, allowing for translation. For additional context: AutoEncoders are also frequently adversarially regularized so you&amp;#x27;re not wrong about GANs being involved. My understanding is that the most accurate implementations also have some manual feature engineering for more tractable problems like facial orientation. One of my favorite related papers: https://arxiv.org/pdf/1703.00848.pdf&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1703.00848/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CV</category></item><item><title>Unsupervised Image-to-Image Translation Networks — mentioned by psyc</title><link>https://news.ycombinator.com/item?id=16246384</link><guid isPermaLink="false">urn:hackerpapers:mention:1703.00848:16246384</guid><pubDate>Sat, 27 Jan 2018 14:49:10 GMT</pubDate><description>&lt;p&gt;The reddit user who wrote the script said it was at least loosely based on: https://arxiv.org/abs/1703.00848&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1703.00848/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CV</category></item><item><title>Unsupervised Image-to-Image Translation Networks — mentioned by clickok</title><link>https://news.ycombinator.com/item?id=15901649</link><guid isPermaLink="false">urn:hackerpapers:mention:1703.00848:15901649</guid><pubDate>Mon, 11 Dec 2017 23:16:46 GMT</pubDate><description>&lt;p&gt;I am disappointed in the level of fear-mongering pearl clutching; it could[0] get so much worse! First, the person creating the fakes is using an off-the-shelf algorithm[1], adapted accordingly. A system designed purely for the purpose of swapping faces (or morphing a particular face into others, one-to-many style) would obviously yield improvements. Secondly, it&amp;#x27;s the work of one person without commercial backing[2]. A studio that invested in lidar or used multiple cameras would be able to incorporate volumetric information. Third, it&amp;#x27;s limited to individuals where you have a performer…&lt;/p&gt;&lt;p&gt;&lt;a href="https://hackerpaper.kaipfister.de/p/1703.00848/"&gt;Paper and discussion history&lt;/a&gt;. A mention is not an endorsement.&lt;/p&gt;</description><category>cs.CV</category></item><item><title>Unsupervised Image-to-Image Translation Networks — HN submission</title><link>https://news.ycombinator.com/item?id=13794753</link><guid isPermaLink="false">urn:hackerpapers:submission:1703.00848:13794753</guid><pubDate>Sun, 05 Mar 2017 05:39:38 GMT</pubDate><description>&lt;p&gt;Unsupervised Image-To-Image Translation Networks&lt;/p&gt;&lt;p&gt;&lt;/p&gt;&lt;p&gt;3 points; 0 discussion replies on HN. &lt;a href="https://hackerpaper.kaipfister.de/p/1703.00848/"&gt;Paper and source history&lt;/a&gt;. A submission is not an endorsement.&lt;/p&gt;</description><category>cs.CV</category></item></channel></rss>
