AI generated video

[Mon0Zer0]

Meta (facebook) today announced a new AI that generates video with some stunning results, equivalent to Dall-e2 but ... video.

Make-A-Video by Meta AI

A state-of-the-art AI system generates high-quality videos from text prompts

makeavideo.studio

Here a teddy bear is painting a self portrait.

In the open source world you also have Phenaki - check out some of the video here:

Phenaki

It looks like Phenaki can also make long form videos such as this one:

 

[Astro Pen]

Here is a fun video. I feel some of it's reasoning is deeply flawed, particularly the anthropomorphic errors of robots resentment and anger but worth a watch.

 

[BAYLOR]

I can imagine an Oscar show years down the line where an A I program wins the Academy award for Best Director .

 

[apocalypsegal]

For me is very interesting how AI works in the video editing world.

 

[Robert Zwilling]

Tyler Perry Halts $800 Million Expansion which would have added 12 more sound stages to his Atlanta Studio after trying out text to video service. Thinking is that filming on location for making movies could eventually become a thing of the past. Viewing the world through a video lens is becoming a common place scenario. Adjusting the picture, and reality, is rapidly becoming child's play.

 

[apocalypsegal]

I've been in video editing for a long while, and the whole concept of AI generating video still blows my mind. It's hard for me to wrap my head around how AI can play a role in creating content that usually requires a human touch. I even gave AI tools a shot to see if they could help streamline my editing process, but in the end, I found myself going back to using Movavi. It's just more straightforward for what I need, especially when it comes to editing and then needing to video compressor for whatsapp to share projects with clients and team members directly. I rely heavily on WhatsApp for work communications, and Movavi has been a reliable tool for that. So, despite the advancements in AI, I'm sticking with what works best for me.

 

[paeng]

 

[Rodders]

My friend has been playing with a lyric and music generating AI called Suno. It's very clever.

 

[paeng]

From a few years back:

 

[KGeo777]

The Cruise video was amazing--however after a few viewings the fakeness becomes apparent.
Longevity of the illusion seems to be a particular hurdle with the technology.
CGI dates so quickly.

There is an Elvis performance which was pretty amazing since it took a while to realize it couldn't be real as the video looked modern and had no film grain.
I showed it to some people and they asked why I was showing them Elvis.
I am not overly familiar with his singing antics so I don't know if there is any slip ups but this is the best Deep Fake I have seen:

 

[paeng]

Also, The Irishman

 

[Harpo]

I look forward to more weirdness like this

 

[RDArmstrong]

Pretty convincing

 

[Christine Wheelwright]

AI seems better at paws than hands.

 

[Ursa major]

AI seems better at paws than hands.

That's because video equipment and software have always had a pause button... oh, and using AI is supposed to be a hands-off process....

 

[Omits]

Meta (facebook) today announced a new AI that generates video with some stunning results, equivalent to Dall-e2 but ... video.

Make-A-Video by Meta AI

A state-of-the-art AI system generates high-quality videos from text prompts

makeavideo.studio

Here a teddy bear is painting a self portrait.
In the open source world you also have Phenaki - check out some of the video here:

Phenaki

It looks like Phenaki can also make long form videos such as this one:

Can AI detect a Deepfake?

 

[Harpo]

My new favourite AI stuff is cinematic reimaginings such as this

 

[Harpo]

 

[Harpo]