Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
1 changed file
with
18 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,18 @@ | ||
--- | ||
title: "Avatarify" | ||
date: 2021-01-09 | ||
tags: ["hack"] | ||
author: sweemeng | ||
image: https://res.cloudinary.com/dty81dwqf/image/upload/c_scale,w_872/v1609732172/faking_einstein_ommi4n.jpg | ||
--- | ||
![](https://res.cloudinary.com/dty81dwqf/image/upload/c_scale,w_872/v1609732172/faking_einstein_ommi4n.jpg) | ||
|
||
Recently I discovered a thing called [First Order Motion Model](https://github.com/AliaksandrSiarohin/first-order-model). This is a technique to create deepfakes using a driver video and one image. It have a few interesting use, other than deepfake. One particular one is testing clothes. | ||
|
||
But today I am doing something more mischievous. I am doing this to pretend to be someone on webcam in a chat. The way I do this is [Avatarify](https://github.com/alievk/avatarify#linux). This is an application running on a computer that create a virtual webcam for you to pretend as someone else. In this case I pretend to be einstein and Mona Lisa. | ||
|
||
![](https://res.cloudinary.com/dty81dwqf/image/upload/c_scale,w_997/v1609732884/Screenshot_from_2020-12-24_21-43-54_emkydk.png) | ||
|
||
The setup step on the page just works for me on linux running on a GTX1060. It requires you to install Anaconda to work. One very interesting scenario is the use of Google Colab to drive the image. I didn't try it, it worth to do so if you don't have a powerful graphics cards. | ||
|
||
Arguably using a real human face is a little evil. But you can use use paintings for [example](https://www.reddit.com/r/MotionDesign/comments/j61scc/r_first_order_motion_model_applied_to_animate/). I can foresee many fun we can use for this. One day even animate cartoon or anime characters. Just don't evil. |