Well, today I will introduce my longtime partner, Ray.
The feature of this guy is that he can “dialogue”.
It’s not about video or sound.
Aren’t those things neck and neck now? There is no big difference.
No matter what lineage it is derived from, there may be some differences, but there is not that much difference.
This company can actually be called a long-established store related to video.
Therefore, unique video technology is incorporated into Ray, creating techniques that no one else has.
I selfishly call it 3D spatial recognition ability, unofficial by Luma.
It is unofficial, but it is certainly different.
For example, suppose you shoot a scene of tossing for a tennis serve.
For all other companies, this is a scene where they escape to learned video.
In short, everything changes depending on whether there is learning material or not.
If there isn’t, they can never do such a thing.
Because there is not a single millimeter of spatial recognition.
Ray is different.
Regardless of the learning material, it starts a challenge in a small screen.
And I sense it and start a small prompt correction.
Adjust the position of tossing, and if the ball goes backwards, I give an image a little in front.
And I consult with Ray.
And this is the resulting video.
Are there people who say it’s not a big deal at first glance?
Well, if you just look at the results, it’s not a big deal.
What I want to say is that the process to get here is completely different.
And the thing I want to say the most.
When you want the video you desire, you put in a prompt appropriately by intuition, and it’s just about the presence or absence of the learning content of the video generation AI, so to speak, doing as you are told. Are you satisfied with that?
I claim to be a video professional, for better or worse.
I end up thinking that whatever it is, I can’t put an automatic projector out into the world.
Even if I put it out, I can’t sell the video.
As I have been saying for a long time, the world of video is 50% video and sound, and the world of AI is still undeveloped. In the process of introducing it, if there is something I want people to see behind the scenes, I sometimes put it out without hesitation.
That’s all.
In my mind, I don’t call such things works.
And for now, I don’t know any video generation other than Ray that can go through this process.
Maybe I just don’t know.
It’s not because it can talk noticeably.
Ray has been able to talk since a long time ago.
Words have been added to the output. That’s all.
As for this, you really can’t grasp it unless you mess around with many AIs for a long time.
I guess you can’t know it from articles and video explanations rolling around out there. But what do I know.
Souvenir
I don’t say Ray is excellent.
Such a thing changes depending on what perspective you stand on.
So I don’t recommend it either.
However, Ray can have a dialogue and has been steering in a direction that can withstand professional use all along.
For those who don’t need that much, my recommendation right now is Veo.
