In this post, I will state my approach for exploring Deep Fake.
Resource
There are several codes that replace faces in videos.
The two most important resources:
- https://github.com/AliaksandrSiarohin/first-order-model
- https://github.com/iperov/DeepFaceLab
There are several blogs that test their effectiveness.
- https://towardsdatascience.com/how-to-produce-a-deepfake-video-in-5-minutes-513984fd24b6
- https://jsoverson.medium.com/from-zero-to-deepfake-310551e59aa3
- https://medium.com/analytics-vidhya/building-deep-fake-videos-cbdabc00e4c1
Right now, I tried the first order model. It was not working well on lab videos. I will think some more and update accordingly. Seeing the effectiveness of this code as tested by several blogs, I believe this will work. Just, I need to figure out what am I missing.