Seeing is believing. Human beings have grown and evolved in an environment where our eyes play a fundamental role, that of identifying what we are interacting with. Reality is created primarily by vision, and that makes us give it critical importance, even enough to trust what we see almost without hesitation. And this is already a problem even in the workplace.
Already in 1938, Orson Welles made an impact with his radio work War of the WorldsWhat made thousands believe we were facing an alien invasion, without having to display anything visually. Imagine if you had access to today’s technology where you could display whatever you wanted, just by using applications that we carry on our mobiles.
Although we’ve gotten used to seeing digital edits in series and movies by well-known actors, we still haven’t gotten used to the idea that the necessary programs for this can be used even from a mobile, although at the moment they do not have the same quality as those used in the audiovisual industry, of course. But it has already made the leap into advertising a few years ago, with the famous Lola Flores ad. for a Spanish company.
Examples like this made it clear to us that it was something that could be done and people could be fooled into creating these elaborate videos, but the tools, hardware and software, weren’t up to all . They still aren’t, but something is starting to change. Now we can do it from home
now i’m famous
We are entering a phase of technology development in which we can begin to question what we see through a screen. It doesn’t matter if we are watching a person doing something in front of us, on a mobile or on a monitor, even if it looks real, it may not be. The thing is, the tools needed for this are starting to make the leap to the mainstream.
To see how far you can go without having technical knowledge of the creation wrong wrong we decided to test one of the tools that allow it. It’s called Unboring and it allows change my face from one video to another by simply extracting it from a photographas you can see in the video above.
The results are not, for the moment, sufficient to deceive, but it seems like a matter of time before the technique evolves and we are able to create realistic videos with web tools. However, there are some things you can do to refine the final result as much as possible.
three tips
After trying dozens of different celebrity videos and faces, I realized it’s important to have three things clear. To begin with, it is important choose a person who has the same facial structure as us. As you can see in the videos, actors with a beard and a wider or bigger face look better in my case, because they fit my own structure. If I didn’t have a beard or had a thin face, I would have chosen other examples. If you’ve ever been told that If you look like a particular person, you can start there.
you also have to choose a video with good lighting, even better if it is recorded on purpose for him, with the background that we prefer, the framing and even the audio, if we are going to use it. Keep in mind that the sound will not change.
Finally, it must be taken into account that the web does not change the hairstyle, so it is better to choose people who have the same style or, at least, who have had it in the past. I, for example, didn’t catch any actor bald or shaved for obvious reasons.
how to change face
Knowing all we can go on the web and register us, for which we must link our Google or Facebook account. When we do, we just have to choose two things, the video we want to upload, on the left, and the face we want to appear thereon the right.
When we did this we will have to spend one of the 5 tokens they give us when we register, and if we want more, we will have to pay or create a new account. At the bottom we can switch between the original and the new video, as well as improve the quality, although this function is still under development. It’s a shame because the difference in quality between a video recorded with a mobile phone and that which appears on the web is quite noticeable.
AI goes much further
But the applications of this technology go beyond just pretending to be someone else. Services like Synthesia, Elai or Rephrase, allow generate explainer videos, perfect for business or educational environments, especially for online courses in which the teacher often registers to download the audiovisual material that the students can use at any time. You can choose not only the virtual avatar, but also the language in which you want the video to be generated.
They can also be used to create artistic videos. That’s where services like Kaiber come in, giving a look concrete to the climbed parts. This type of use is perhaps the least problematic, at least politically and socially, but it will undoubtedly change the way content is created. New artificial intelligence applications are also capable of create videos from text commands, as is the case with applications like Midjourney for photos. There are services that allow you to create videos from scratch, like Gen2, although the quality is still very limited. It’s a matter of time before its performance increases.
It is clear that society is entering a new era in which even what is seen on video cannot be taken as a reliable example that it is true. And it is also a problem in the opposite direction, because there will be many who want to avoid responsibility even after being caught on camera on the pretext that in reality it is not them, even if they look like them.
You may be interested
Follow the topics that interest you
Table of Contents