I was watching this video from corridor crew, according to them, they have used deepfake technology to create this video. I myself have never made a deepfake videos, but I have enough knowledge in the underlying technology to know that it's hard to swap a face with multiple people existing simultaneously in a frame and let alone swapping faces of multiple people in a single frame. But corridor crews video showed that multiple deepfakes can be done, and that's why I am sceptical about the video of using deepfake technology.
If the video is indeed made with deepfake technology, then what is the mechanism behind this? My own guess is that they might have masked other people and concentrated on one in a frame. Then they have used this masked frame to generate the deepfake, which then concatenated to the original frame. Do you think this is possible? Is there a research article or blog post which explains this process?
