Sneak Peek at AI Text to Video Generation - RunwayML Gen 2

RunwayML Gen 2 is a multi-modal AI system that can generate novel videos with text, images, or video clips. It is a breakthrough for generative AI and an in-house research effort by Runway Research. It is the first publicly available text-to-video generator and it can also transfer the style of any image or prompt to every frame of a video, turn mockups into fully stylized and animated renders, isolate subjects in a video and modify them with simple text prompts, and more. Today, we get to see a sneak peak!

▼ Link(s) From Today’s Video:

-------------------------------------------------

▼ Extra Links of Interest:

-------------------------------------------------

Thanks for watching Matt Video Productions! I make all sorts of videos here on Youtube! Technology, Tutorials, and Reviews! Enjoy Your stay here, and subscribe!

All Suggestions, Thoughts And Comments Are Greatly Appreciated… Because I Actually Read Them.

-------------------------------------------------

[xules]

It will be interesting when you can feed a book or even an audio book in to one of these generators to get a movie.

[VperVendetta]

Wow that was amazing. I didn't know you could already create consistent characters in different video shots. That short movie was honestly really cool considering we are at the beginning of this. It might even go down in history as the first AI generated movie. Crazy times.

[kompst_tu]

It's crazy because the ai is so advanced and constantly improving itself that within a week it's 50% better

[emmasnow]

What I love about this is how amazing this will be by the end of the year.

[wesleytillman]

This technology is heading toward a supernova of creative possibilities that will redefine our entire culture. The power of Hollywood level movie creation will before long be in the hands of anyone who wants to try their hand at movie-making with a capacity of what used to cost a fortune and would employ a movie crew the size of a small army.

[FusionDeveloper]

Cool. I am still waiting to see unbounded length.
I would love to have this level of coherency to be able to make a 10-15 minute long video from 1 prompt.
I've tried using other things before and couldn't figure out a way to "continue" from the end of the last clip because it completely changes everything.

[pladselsker]

I think I'm still gonna continue doing still images even when video models gets released. A little like when the camera first was invented, people thought it would kill the photograph, but it never did. I think still images are just a different way to express something.

With videos, each frame may not all stand well on its own, but as a sequence of frames it can make a masterpiece.

Images, in contrast, stand on their own, but only capture a single moment.

I'm sure they both have pros and cons. They will have a place depending on what you're trying to do.

[ArjanTigchelaar]

That short film was amazing! Actually quite chilling.

[GabrieloTomorrow]

Wow...

I was predicting that we could make AI generated films in about 5 years, and that in 2023 we were probably going to see the first steps of text-to-video AI becoming mainstream.
But things are going a LOT faster than I was expecting.

Now I'm actually starting to believe that by the end of this year we'll already be able to make some videos almost indistinguishable from movies that were shot with cameras.

This is so crazy and exciting!

[arnelilleseter]

I can't wait to see what people will create with this in the future. No restrictions from corporations who only care about what is most popular at the moment, no more worrying about budget limits.
There are so many stories just waiting to be told.

[revan]

Do you think we will eventually be able to generate new seasons of cancelled TV shows?

[Aldraz]

Yeah, videos are definitely gonna replace most images, because you can make "still" videos in 90% of the cases, like imagine a 3D photo that moves just tiny bit, like wind moving hair, or raining in the background.. yeah I can see it much more popular than images for sure.

[johno]

this is the worst this technology will ever be

[TheGamersSanctuaryNetwork]

The pacing is definitely good! I can’t wait to see how good these technologies get in the very near future. Makes me wonder how the labor market will look in 20 yrs.

[promptengineer]

This is so cool! I can't wait to try this new text to video technology, I can't wait for the beta to open up to more people.

[NathanLorenzana]

Oh my, I'm experiencing all over again the same awe I had when first playing with Stable Diffusion in Discord. That short film gave me A Scanner Darkly vibe.

[CATDHD]

I am overwhelmed with ai tools. I decided to make a short within a week. Thanks, pal

[bambinoesu]

I can't wait when this releases, about to unleash my creativity to the core

[MaJetiGizzle]

The future of movie generation is closer than ever before.

[rm]

Things I liked a bit cuz I could use it to visualize what I want for my Art creation or film making.
Everything does not have to be perfectly well-done.

This level is still amazing to help us.

0:50

2:20

13:47

15:01

17:35 good one to get a idea of when you writing a script or something.

18:10 Midjurney 2 Gen2 Video

19:30~23:04 Main Video.