Samsung AI technology will allow people to create videos from a single picture, these videos are also called DeepFake videos.
As deep fake videos are getting popular, Samsung has stepped into the ship with their research that will allow a single static picture into a moving video.
As per the report from Motherboard, Researchers in Samsung AI center in Russian city Moscow has achieved this by coaching a “deep convolution network” through this way of showing a number of videos of talking faces, to allow it and spot and contain the information of specific facial expressions and features than using the same AI to animate the picture into a video.
Here is how Samsung AI technology is reshaping the future by making video our of any picture.
“Few-Shot Adversarial Learning of Realistic Neural Talking Head Models.” is the research paper where the researchers have given the results and process. Unlike other deep fake videos, which needs a number of pictures to animate the picture, Samsung’s AI would only require a single picture to animate it, making it more efficient than other platforms.
In the video, Samsung used the images of Salvador Dali, Albert Einstein, Marilyn Monroe, Mona Lisa, and Fyodor Dostoevsky and turned them into a video and they looked real enough to convince anyone that it is them who are talking. But it will be easy for a person who knows how AI works will able to spot if it is real or not, but most of us will be fooled by it.
The results of the research are worrisome as this technology will allow anyone to create a fake video of someone else speaking the words they have never before. Making people’s lives more difficult as this can also be used to spread fake news, and after seeing its clarity, it will soon become impossible to spot which is real and which is not. We hope this new AI technology does not become a burden on technology users all around the world.
