For several years on the internet there are so-called deepfake videos, but in recent months its popularity has increased greatly thanks to various tools and applications that make it easier to create.
Deepfake is a technique used to modify images or videos using artificial intelligence, in most cases it is used to superimpose one face on another, thus creating fake videos with impressive results.
Before, creating deepfake videos was a skill that took years to learn, but now, with the advancement of technology, it is very easy to do it.
DeepFaceLab is a tool available for Linux and other systems that allows you to create deepfake videos in a very simple way using the command line.
It may seem that using DeepFaceLab is very difficult to use, but nothing far from the opposite, just by running a few scripts you can modify a video and add a new face to the protagonist.
How to install DeepFaceLab on Linux
To install DeepFaceLab you will have to first install Anaconda3 using the Official Site and then you will have to initialize it with the following commands:
export PATH = ~ / anaconda3 / bin: $ PATH conda init bash
Now you can proceed to the installation of DeepFacebLab with the following lines of code:
conda create -y -n deepfacelab python = 3.6.6 cudatoolkit = 9.0 cudnn = 7.3.1 conda activate deepfacelab git clone https://github.com/lbfs/DeepFaceLab_Linux.git cd DeepFaceLab_Linux python -m pip install -r requirements-cuda .txt
If you have Ubuntu 16.04 or 18.04 you can check the official build page for ways to install DeepFaceLab on both systems.
Using DeepFaceLab can be tricky but once you read this always up-to-date tutorial you will be able to create your deepfake videos like a pro.
If you have any problems with the installation or use of DeepFaceLab, we will gladly help you in the comments.