ITCooky Recipes

In the first release of the official model for Stable Diffusion, they forgot to remove the real people there, of course we weren’t there, but celebrities were there. Later they were removed, although it seemed that no one complained… But there is a way to add your own face and poses, and its a great benefit for households, to create their own avatars with any state of the face or stylish background, and for fans of all types: to put your favorite actor in any role, never seen, even impossible.

I assume you already have Stable Diffusion installed via AUTOMATIC1111 web, on some platform, if not, this is how I installed it on UbuntuHow to generate our images using the Stable Diffusion neural network with the textual description!

Today neural networks have made a lot of noise, in one state, as they say, the neural network generated for a woman the image of her deceased husband, so real that she felt him and even get pregnant.

The enthusiasm for neural networks is understandable – the quality of the result scares … artists and other authors. And for us, who are not artists, the neural network simply provides a ready tool for the realization of our ideas: the camera is also a tool for the realization of our ideas, but artists do not ask to prohibit or limit it! (not anymore, at first maybe, yes)!

As long as you live you must reinstall Asterisk – every time like the first time, solving problems for half a day, I don’t know why that happens! While the video card is training the neural network… Yes, and Orange PC Plus, by the way, from the first article Let’s beat spam calls via PSTN line with Asterisk with IVR by Grandstream ht813 on OrangePi PC Plus! – eMMC died, and the whole system… It’s been a long time since I’ve done anything with Raspberry Pi, the cheapest PC, ooh! So what is happening, why in China the 4Gb version now costs $ 170, and in our country about 250 (ah andOrangePi PC Plus is already worth $40!!!)!!! Well, at least the cases for the RPi haven’t gone up in price, I took the Argon One, the version they no longer produce, I still have many accessories for the raspberry here, that’s the current version with two big HDMI Argon ONE V2 Case for Raspberry Pi 4 for just $25, that much you used to pay for RPi in old times!

The usual everyday situation – an old file, a forgotten passwords, what can be done? It can be found by trying all possible passwords in an appropriate time using GPU, of course, if it is a human password and not 8 characters with a lowercase letter, a uppercase letter, a Cyrillic, an English, a number, a special character! You can also do it on the CPU and not on the GPU, but instead of months it will take years, it’s all about the cores in the processors, there are only a dozen cores in the CPU and there are thousands in the video card.

The unforeseen happened, the creators of the most popular Linux for making websites sent them all to hell and they did it some time ago, it’s just now that I felt it myself when I wanted to update something and neither repositories work. The problem with CentOS was its brotherhood with Red Hat – commercial Linux, they pulled the code from the community (returning to CentOS), but at some point vendors said “fuck everyone”, that was the end of CentOS. And another thing – Hetzner prices go up 10% from September for new customers, but for old ones only from 1 January… Everything, everything, everything is pro to migrate now, and there is only one Ubuntu option left!!!

The initial idea was this: the Telegram bot saves all the images that are sent to it, then, every few minutes, it starts processing the images in the folder and the result is sent to everyone. The name of the recipient will be in the name of the file… everything seems to be simple, but it turned out to be very difficult to do: you have to go deeper into asynchrony and, in general… Soo everything will be simple and linear, bot receive a photo, go to process it immediately and let the others wait! And this time I will do it in Python, not in vain I heard (and well, I did all the homework) a free course CS50’s Introduction to Programming with Python, and last time (4 years ago Mamma Mia!) everything was done in node.js Smart house monitoring with TelegramBot in Node.js!

Why do I need a professional card? And that’s the right question…

I have been watching this card since the beginning of the anomalous strengthening of the ruble and managed to buy it before its expected crash! I found a store where the price in dollars corresponded to Westerners (well, almost + 15-20% normal, taking into account the logistics and fat of the store), and in rubles corresponded to the course of the Russian stock market and bought it, thanks www.regard.ru!

At first I looked at A4000 and here is a comparison with A4500 and A5000

Another video effects project with neuroprocessing STIT – Stitch it in Time can rejuvenate the face, age, change sex and much more, but with a but!

First the disadvantages noted:
– Poorly described, nothing at all except a couple of examples.
– It doesn’t make a full video with sound, it will show some kind of demo without sound (can be fixed)
– Cannot choose a face if there are many in the frame (cannot be corrected)
– You can’t select only a face. exmaple if you put your hand on it

GFPGAN is a set of tools and a ready-to-use neural network (you can train your own) to enhance faces in photographs, they also promise to improve the no face places, but somehow I did not notice the enhancement effect, although they resize it well! Project page: github.com/TencentARC/GFPGAN. Everything works on Linux, of course, and it works fine, but today I’m going to use it on Windows, simply because Windows already has Linux called WSL!

I`ll need:
– Windows 10-11 latest version
– Nvidia video card with CUDA (I have rtx 2070), but it can also be done with only CPU!
– CPU with virtualization (almost all CPUs have it)

I finally found a solution specifically to improve blurry photos while preserving the original face, and it’s amazing!

First of all thanks a lot to Tao Yang, Peiran Ren, Xuansong Xie and Lei Zhang for GAN Prior Embedded Network for Blind Face Restoration in the Wild
Entregan gratis la versión pública del proyecto en github github.com/yangxy/GPEN. GPEN allows you to quadruple faces, it could quadruple many faces on same image. Like all great programmers, they don’t know how to write installation and usage instructions, that’s why I’m here …