Let’s face it – all the big data, deep learning, neural network, computer vision – they got the recognition of the regular internet crowd after the instant boom of deepfakes since last few years. The term ‘Deepfake’ itself was coined as an amalgamation of ‘deep learning’ and ‘fake’.
The actual concept of these deepfakes are nothing new. Movies like Face/Off or Mission Impossible popularized the idea of face swapping, but what if you can do the same with your regular PC, some videos of the subject and some free time?
DeepNude ( @deepnudeapp) is the recent internet phenomenon, which can be thought of as the spiritual (voyeurish?) successor to the deepfakes. As you can speculate from the name, the app can remove the clothings from a picture of woman with the eternal power of artificial intelligence!
According to the creator, the app works on the basis of the Pix2Pix algorithm developed by researchers from the University of California, Berkeley. Both Windows and Linux compatible binaries were published (the links are now offline), and they could work without internet after downloading required library files (~2 GB).
The app was actually free, but you had to spend a one time fee of $50 for the ability to get rid of the watermark and export the processed pictures. Although the developer disables the download links, malware infested copies are circulating through various torrent trackers and social medias.
Wait, there’s more! Some enthusiasts have already reverse engineered the app and bypassed the in-app purchase mechanism, so that end users can use the cracked copies of the app. The resultant images are far from being ‘perfect’, but they are enough to harass people.
Besides obscure forums and communities, the copies of the program can be found in various public platforms like Telegram/Discord channels, subreddits as well as warez sites. The situation is similar to the loli crackdown by reddit mods a few months ago, but the intensity is more prominent this time.
While it is still possible to download untouched copies of the app from third party mirrors and torrents, PiunikaWeb advises to stay away from them. You may lose more than you can get!
Follow @PiunikaWeb
PiunikaWeb is a unique initiative that mainly focuses on investigative journalism. This means we do a lot of hard work to come up with news stories that are either ‘exclusive,’ ‘breaking,’ or ‘curated’ in nature. Perhaps that’s the reason our work has been picked by the likes of Forbes, Foxnews, Gizmodo, TechCrunch, Engadget, The Verge, Macrumors, and more. Do take a tour of our website to get a feel of our work. And if you like what we do, stay connected with us on Twitter (@ PiunikaWeb) and other social media channels to receive timely updates on stories we publish.
Tags
deepfake deepnude
Kingshuk De
896 Posts
I came from a mixed background of Statistics and Computer Science. My research domains included embedded computer systems, mobile computing and delay tolerant networks in post-disaster scenarios. Apart from tinkering with gadgets or building hackintosh, I like to hop on various subreddits and forums like MyDigitalLife and XDA.
Next article View Article
