github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

  • A+

最近这几天AI福利应用DeepNude突然火了,该软件竟然可以一键直接“脱掉”女性的衣服,然后迅速火爆全球。小伙伴们纷?#36164;誅eepNude种子下载,其实这个DeepNude软件在网站上不难找,不信你在艾薇这里搜索一下有惊喜

DeepNude应用也是非常的容?#36164;?#29992;,只需要给DeepNude导入一张照片,即可借助AI神经网络技术,DeepNude自动“脱掉”图片里的衣服。DeepNude原理虽然理解门槛高,但是应用起来却毫不费力,因为对于使用者来说,DeepNude无需任何技术知识,一键即可获取。这才是火的原因,不然技术含量太高的话使用的人会大大的减少的。

据DeepNude开发者表示,DeepNude研发团队是个几个人的团队,而且相关技术显然?#19981;?#24456;不成熟,使用的是开源的算法,而且多数照片(尤其是低分辨?#25910;?#29255;)经过DeepNude处理后,得出的图像会有?#26031;?#30165;迹;只有女性照片效果好一点,而在DeepNude输入卡通人物照片,生成得出的图像是完全扭曲的,DeepNude大多数图像和低分辨率图像会产生一些视觉伪像。

 

如果DeepNude导入目标图片还是各种女性,最早曝光这一应用的科技媒体Motherboard表示,他们通过100多张的照片测试,发现如果输入《体育画报泳装特辑》(Sports Illustrated Swimsuit)的照片,得到的果体照片最为逼真。可能是因为果露的部分非常的多吧。

 

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

 

DeepNude应用瞬间引发了网络社区的各类声讨,纷纷明确的表示是对AI利用的反例。

 

DeepNude应用在一片讨伐声中很快下线,但是DeepNude的影响力还在。特别是对DeepNude应用背后技术的探讨还一直在?#20013;?/p>

 

最近一个名为“研究DeepNude使用的图像生成和图像修复相关的技术和论文“的GitHub升至一周热榜,获得了不少星标。

 

项?#30475;词?#20154;四川大学计算机学院在?#20102;?#22763;袁宵显然对于这一项目背后的技术很有研究,袁宵提出了其生成需要的一系列技术框架,以及哪些技术可能有更好的实?#20013;?#26524;。艾薇在此进行转载,希望各位技术专?#20197;?#28385;足技术好奇心的同时,也可以正确使用自己?#31181;?#30340;技术力量。

Research techniques and papers related to image generation and image restoration used by DeepNude. 研究DeepNude使用的图像生成和图像修复相关的技术和论文。

DeepNude-an-Image-to-Image-technology
Reprinted to indicate the source 转载注明出处 :
https://github.com/yuanxiaosc/DeepNude-an-Image-to-Image-technology

作者:四川大学计算机学院在?#20102;?#22763;袁宵GitHub

Next I will open up some image/text/random-to-image neural network models and utilities for learning and communication, and also welcome to share your technical solutions.

接下来?#19968;?#24320;源一些image/text/random-to-image的神经网络模型和实用工具,仅供学习交流之用,?#19981;?#36814;分享你的技术解决方案。

Image-to-Image Demo 图像到图像demo

https://affinelayer.com/pixsrv/

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)
Stick figure to colorful cats/shoes/handbags demo 简笔画到色彩丰富的猫、鞋、手袋 demo

DeepNude software mainly uses Image-to-Image technology, which theoretically converts the images you enter into any image you want. You can experience Image-to-Image technology in your browser by clicking Image-to-Image Demo below.

DeepNude 软件主要使用了Image-to-Image技术,该技术理论上可以把你输入的图片转换成任何你想要的图片。你可以点击下方的Image-to-Image Demo在浏览器中体验Image-to-Image技术。

Image-to-Image Demo

An example of using this demo is as follows:

In the left side box, draw a cat as you imagine, and then click the pix 2 pix button, you can output a model generated cat.

在左侧框中按照自己想象画一个简笔画的猫,再点击pix2pix按钮,就能输出一个模型生成的猫。

DeepNude Technology 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.

这一部分阐述DeepNude相关的?#26031;?#26234;能/深度学习理论(特别是计算机视觉)研究,如果你?#19981;对?#35835;论文使用最新论文成果,尽情享用吧。

1. Image Inpainting 图像修复
论文 NVIDIA 2018 paper Image Inpainting for Irregular Holes Using Partial Convolutions and Partial Convolution based Padding.

https://arxiv.org/abs/1804.07723

代码 Paper code partialconv。

https://github.com/NVIDIA/partialconv

效果

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

In the image interface of Image_Inpainting(NVIDIA_2018).mp4 video, you only need to use tools to simply smear the unwanted content in the image. Even if the shape is very irregular, NVIDIA's model can “restore” the image with very realistic The picture fills the smeared blank. It can be described as a one-click P picture, and "no ps traces." The study was based on a team from Nvidia's Guilin Liu et al. who published a deep learning method that could edit images or reconstruct corrupted images, even if the images were worn or lost pixels. This is the current 2018 state-of-the-art approach.

在 Image_Inpainting(NVIDIA_2018).mp4 视频中左侧的操作界面,只需用工具将图像中不需要的内容简单涂抹掉,哪怕形状很不规则,NVIDIA的模型能够将图像“复原?#20445;?#29992;非常逼真的画面填补被涂抹的空白。可谓是一键P图,而且“毫无ps痕迹”。 该研究来自Nvidia的Guilin Liu等人的团队,他们发布了一种可以编辑图像或重建已损坏图像的深度学习方法,即使图像穿了个洞或丢失了像素。这是目前2018 state-of-the-art的方法。

2. Image-to-Image or Pix2Pix (need for paired train data)
DeepNude mainly uses this Image-to-Image(Pix2Pix) technology.

论文 Berkeley 2017 paper Image-to-Image Translation with Conditional Adversarial Networks.

https://arxiv.org/abs/1611.07004

主页 homepage Image-to-Image Translation with Conditional Adversarial Nets

https://phillipi.github.io/pix2pix/

代码 code pix2pix

https://github.com/phillipi/pix2pix

Run in Google Colab pix2pix.ipynb

https://github.com/tensorflow/docs/blob/master/site/en/r2/tutorials/generative/pix2pix.ipynb

效果

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

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.

https://arxiv.org/abs/1611.07004

Image-to-Image Translation with Conditional Adversarial Networks 是伯克利大学研究提出的使用条件对抗网络作为图像到图像转换问题的通用解决方案。

3. CycleGAN (without the need for paired train data)
论文 Berkeley 2017 paper Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
https://arxiv.org/abs/1703.10593
代码 code CycleGAN
https://github.com/junyanz/CycleGAN

Run in Google Colab cyclegan.ipynb
https://github.com/tensorflow/docs/blob/master/site/en/r2/tutorials/generative/cyclegan.ipynb

效果

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

CycleGAN uses a cycle consistency loss to enable training without the need for paired data. In other words, it can translate from one domain to another without a one-to-one mapping between the source and target domain. This opens up the possibility to do a lot of interesting tasks like photo-enhancement, image colorization, style transfer, etc. All you need is the source and the target dataset.

CycleGAN使用循环一致性损失函数来实现训练,而无需配对数据。 换句?#20843;擔?#23427;可?#28304;?#19968;个域转换到另一个域,而无需在源域和目标域之间进行一对一?#25104;洹?这开启了执行许多有趣任务的可能性,例如照片增强,图像着色,样式传输?#21462;?#24744;只需要源和目标数据集。

horse2zebra 马变斑马

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

DeepNude software itself
After researching DeepNude technology, I have removed data related to DeepNude. Please don't ask me to get DeepNude program.

DeepNude's technology stack

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)
Python + PyQt
pytorch
Deep Computer Vision
Windows version of DeepNude use process

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区) github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区) github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区) github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)
DeepNude can really achieve the purpose of Image-to-Image, and the generated image is more realistic.

Delete the color.cp36-win_amd64.pyd file in the deepnude root directory, and then add the color.py file to get the advanced version of deepnude.

What can be improved?
DeepNude software shortcomings

Size. Including 156M DeepNude_Windows_v2.0.0.zip and 1.90G pyqtlib.rar;
Speed. It takes 30 seconds to convert a picture;
Content. Use the Image-to-Image neural network to automatically remove the clothes from women to reveal their nudity. This application applies the wrong application of deep learning.
Where DeepNude can be improved

DeepNude can be implemented using Tensorflow and uses model compression techniques.
DeepNude should change the current practice of not respecting women.
Future
In fact, we don't need Image-to-Image. We can use GANs to generate images directly from random values or generate images from text.

1. Obj-GAN
The new AI technology Obj-GAN developed by Microsoft Research AI understands natural language descriptions, sketches, composite images, and then refines the details based on individual words provided by sketch frames and text. In other words, the network can generate images of the same scene based on textual descriptions that describe everyday scenes.

微软?#26031;?#26234;能研究?#28023;∕icrosoft Research AI)开发的新 AI 技术Obj-GAN可以理解自然语言描述、绘制草图、合成图像,然后根据草图框架和文字提供的个别单词?#23500;?#32454;节。换句?#20843;擔?#36825;个网络可以根据描述日常场景的文字描述生成同样场景的图像。

效果

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

模型

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

2. StoryGAN
Advanced version of the pen: just one sentence, one story, you can generate a picture.

Microsoft's new research proposes a new GAN, ObjGAN, which can generate complex scenes based on textual descriptions. They also proposed another GAN, StoryGAN, which can draw a story. Enter the text of a story and output the "picture book".

进阶版神笔:只需一句话、一个故事,即可生成画面

微软新研究提出新型 GAN——ObjGAN,可根据文字描述生成复?#26144;?#26223;。他们还提出另一个可以画故事的 GAN——StoryGAN,输入一个故事的文本,即可输出「连环画」。

当前最优的文本到图像生成模型可以基于单句描述生成逼真的鸟类图像。然而,文本到图像生成器?#23545;?#19981;止仅对一个句子生成单个图像。给定一个多句段落,生成一系列图像,每个图像对应一个句子,完整地可视化整个故事。

效果

github:DeepNude图片生成AI算法原理及源代码分享(技术讨论区)

Researchers should work to improve human well-being, not to gain income through breaking the law..

The most commonly used image-to-image technology should be Beauty App, so why don't we develop a smarter Beauty Camera?

现在用得最多的Image-to-Image技术应该就是美颜APP了,所以我们为什么不开发一个更加智能的美颜相机呢?

发表评论

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen: