How to find a real Deepnude Source Code?

Here are 9 public repositories on Github matching this topic:

yuanxiaosc / DeepNude-an-Image-to-Image-technology

This repository contains the pix2pixHD algorithms(proposed by NVIDIA) of DeepNude, and more importantly, the general image generation theory and practice behind DeepNude.


zhengyima / DeepNude_NoWatermark_withModel


dreamnettech / dreamtime


dreamnettech / dreampower


Yuagilvy / DeepNudeCLI


redshoga / deepnude4video


Sergeydigl3 / pepe-nude-colab


ieee820 / DeepNude-an-Image-to-Image-technology


2anchao / deepnude_test

🔞 DeepNude Algorithm

DeepNude is a pornographic software that is forbidden by minors. If you are not interested in DeepNude itself, you can skip this section and see the general Image-to-Image theory and practice in the following chapters.

DeepNude_software_itself content:

  1. Official DeepNude Algorithm(Based on Pytorch)
  2. DeepNude software usage process and evaluation of advantages and disadvantages.

👍 NSFW

Recognition and conversion of five types of images [porn, hentai, sexy, natural, drawings]. Correct application of image-to-image technology.

NSFW (Not Safe/Suitable For Work) is a large-scale image dataset containing five categories of images [porn, hentai, sexy, natural, drawings]. Here, CycleGAN is used to convert different types of images, such as porn->natural.

  1. Click to try pornographic image detection Demo
  2. Click Start NSFW Research

Image Generation Theoretical Research

This section describes DeepNude-related AI/Deep Learning theory (especially computer vision) research. If you like to read the paper and use the latest papers, enjoy it.

  1. Click here to systematically understand GAN
  2. Click here to systematically image-to-image-papers

1. Pix2Pix

Result

Image-to-Image Translation with Conditional Adversarial Networks is a general solution for the use of conditional confrontation networks as an image-to-image conversion problem proposed by the University of Berkeley.View more paper studies (Click the black arrow on the left to expand)


Image Generation Practice Research

These models are based on the latest implementation of TensorFlow2.

This section explains DeepNude-related AI/Deep Learning (especially computer vision) code practices, and if you like to experiment, enjoy them.

1. Pix2Pix

Use the Pix2Pix model (Conditional Adversarial Networks) to implement black and white stick figures to color graphics, flat houses to stereoscopic houses and aerial maps to maps.

Click Start Experience 1

2. Pix2PixHD

Under development… First you can use the official implementation

3. CycleGAN

The CycleGAN neural network model is used to realize the four functions of photo style conversion, photo effect enhancement, landscape season change, and object conversion.

Click Start Experience 3

4. DCGAN

DCGAN is used to achieve random number to image generation tasks, such as face generation.

Click Start Experience 4

5. Variational Autoencoder (VAE)

VAE is used to achieve random number to image generation tasks, such as face generation.

Click Start Experience 5

6. Neural style transfer

Use VGG19 to achieve image style migration effects, such as photo changes to oil paintings and comics.

Click Start Experience 6

………………………………………………………………..

If you are a user of PaddlePaddle, you can refer to the paddlepaddle version of the above model image generation model library paddegan.

https://www.vice.com/en/article/8xzjpk/github-removed-open-source-versions-of-deepnude-app-deepfakes

Something to consider:

From Wikipedia: “X-Ray Specs are an American novelty item, purported to allow the user to see through or into solid objects. In reality the glasses merely create an optical illusion; no X-rays are involved. The current paper version is sold under the name “X-Ray Spex”; a similar product is sold under the name “X-Ray Gogs”.”

“X-Ray Specs consist of an outsized pair of glasses with plastic frames and white cardboard “lenses” printed with concentric red circles, and emblazoned with the legend “X-RAY VISION”.

“The “lenses” consist of two layers of cardboard with a small hole about 6 millimetres (0.24 in) in diameter punched through both layers. The user views objects through the holes. A feather is embedded between the layers of each lens. The vanes of the feathers are so close together that light is diffracted, causing the user to receive two slightly offset images. For instance, if viewing a pencil, one would see two offset images of the pencil. Where the images overlap, a darker image is obtained, supposedly giving the illusion that one is seeing the graphite embedded within the body of the pencil. As may be imagined, the illusion is not particularly convincing.

“X-Ray Specs were long advertised with the slogan “See the bones in your hand, see through clothes!” Some versions of the advertisement featured an illustration of a young man using the X-Ray Specs to examine the bones in his hand while a voluptuous woman stood in the background, as though awaiting her turn to be “X-rayed”.

New AI App ‘Deep Nostalgia’ Brings Old Photos To Life

Service from MyHeritage uses deep learning technique to automatically animate faces

An AI-powered service called Deep Nostalgia that animates still photos has become the main character on Twitter this fine Sunday, as people try to create the creepiest fake “video” possible, apparently.

The Deep Nostalgia service, offered by online genealogy company MyHeritage, uses AI licensed from D-ID to create the effect that a still photo is moving. It’s kinda like the iOS Live Photos feature, which adds a few seconds of video to help smartphone photographers find the best shot.

But Deep Nostalgia can take photos from any camera and bring them to “life.” The program uses pre-recorded driver videos of facial movements and applies the one that works best for the still photo in question. Its intended purpose is to allow you to upload photos of deceased loved ones and see them in “action,” which seems like a lovely idea.

Like most “deepfakes” like “deepnude” – the name for the popular use of this technology to map one person’s face on to footage of another – the service is exceptionally good at smoothly animating features and expressions. But it can also struggle to generate data to fill in the “gaps” in what it can see from the source photos, causing a sense of the uncanny.

Users have to sign up for a free account on MyHeritage and then upload a photo. From there the process is automated; the site enhances the image before animating it and creating a gif. The site’s FAQ says it does not provide the photos to any third parties, and on its main page a message reads “photos uploaded without completing signup are automatically deleted to protect your privacy.”

Naturally, the program has become something of a meme-generator on Twitter, with users trying to push the AI to its limit. An archaeologist used photos of ancient statues, and yes they included some with the blank eyes. Sorry in advance for the nightmare fuel (but hiiii there Alexander the Great):

“Some people love the Deep Nostalgia feature and consider it magical, while others find it creepy and dislike it,” MyHeritage says about its technology. “Indeed, the results can be controversial and it’s hard to stay indifferent to this technology. This feature is intended for nostalgic use, that is, to bring beloved ancestors back to life. Our driver videos don’t include speech in order to prevent abuse of this, such as the creation of ‘deep fake’ videos of living people.”

Not every video created with the service is elegantly animated, or even good enough to be unsettling, of course. An animated version of the infamous bust of Ronaldo, for instance, is exactly as distressing as the static version:

And while the automatically produced videos of Deep Nostalgia are not likely to fool anyone into thinking they are real footage, more careful application of the same technology can be very hard to distinguish from reality.

Deep Nostalgia can only handle single headshots and can only animate faces, so you’re not going to be able to reanimate mummies to make it look like they’re walking (hey I wondered, OK?). You can upload five photos for free to the MyHeritage website for Deep Nostalgia treatment, after that you have to register for a paid account.

Last month, a new TikTok account named deeptomcruise racked up millions of views with a series of videos that are, it claims, deepfake versions of the actor talking to camera. The Cruise fakes are so accurate that many programmes designed to recognise manipulated media are unable to spot them.

  1. How does the Deep Nostalgia software illustrate the growing problem of deep fakes?
  2. What are the biggest trends in deep fakes at the moment?
  3. How can users protect themselves against being fooled by deep fakes?

word2vec deep learning

Word2Vec — a baby step in Deep Learning but a giant leap towards Natural Language Processing

The traditional approach to NLP involved a lot of domain knowledge of linguistics itself. Understanding terms such as phonemes and morphemes were pretty standard as there are whole linguistic classes dedicated to their study. Let’s look at how traditional NLP would try to understand the following word.

Let’s say our goal is to gather some information about this word (characterize its sentiment, find its definition, etc). Using our domain knowledge of language, we can break up this word into 3 parts.