AI generated short films

wickedpuppy

Arch-Supremacy Member
Joined
Jun 24, 2005
Messages
11,116
Reaction score
6,752
how many more years do we need to wait for AI to be able to generate japanese documentaries?
Years ?

The whole chatGPT thing came out in late 2022 , less than 2 years ago ....

And we are here now ...

How long more you think ?
 

人造人16 号

Master Member
Joined
May 1, 2024
Messages
4,792
Reaction score
4,147
Years ?

The whole chatGPT thing came out in late 2022 , less than 2 years ago ....

And we are here now ...

How long more you think ?

i want an AI that can generate a full length video from a full text story input by the user.
The user also can input photos to the AI for it to generate them as characters for the video.
So basically, a human write a story and the AI generate a full movie out of it, with the character models based on photos inserted by the human.

For AI to be advanced to this stage, i dont think it can be something achievable this or next year.
But for sure this can happen in future, given enough time for AI to be developed.
 

doogyhatts

Arch-Supremacy Member
Joined
Feb 13, 2018
Messages
13,772
Reaction score
3,952
i want an AI that can generate a full length video from a full text story input by the user.
The user also can input photos to the AI for it to generate them as characters for the video.
So basically, a human write a story and the AI generate a full movie out of it, with the character models based on photos inserted by the human.

For AI to be advanced to this stage, i dont think it can be something achievable this or next year.
But for sure this can happen in future, given enough time for AI to be developed.
The open source solution is coming from Bytedance and Nankai University.
Read the paper too.
https://github.com/HVision-NKU/StoryDiffusion

And yes, I believe it is this year, because for the paper to be acceptable at the conference, they have to complete the pretrained model and release the weights.
 

wickedpuppy

Arch-Supremacy Member
Joined
Jun 24, 2005
Messages
11,116
Reaction score
6,752
i want an AI that can generate a full length video from a full text story input by the user.
The user also can input photos to the AI for it to generate them as characters for the video.
So basically, a human write a story and the AI generate a full movie out of it, with the character models based on photos inserted by the human.

For AI to be advanced to this stage, i dont think it can be something achievable this or next year.
But for sure this can happen in future, given enough time for AI to be developed.
Show us your input for AI to generate a full feature length video ...

Or you are going to say a name of a book ?
 

doogyhatts

Arch-Supremacy Member
Joined
Feb 13, 2018
Messages
13,772
Reaction score
3,952
The background is the stage of a theatre.
The noise filter used is subtle.
 

RyanJ

Great Supremacy Member
Joined
Jun 18, 2015
Messages
70,956
Reaction score
69,612
It is probably going to take a while.

The recent impressive 'Air Head' video which OpenAi cunningly did not give much disclaimer except it was made with Sora turned out to require a huge chunk of good old VFX editing, the balloon not being yellow despite prompts etc ...

Something which the production team revealed subsequently.

Also the amount of hardware processing power needed to string a video up is going to be immense and not for the average user any time soon.

 

doogyhatts

Arch-Supremacy Member
Joined
Feb 13, 2018
Messages
13,772
Reaction score
3,952
Very inspiring Japanese YT channel on AI content. Only 3 months old.

 
Important Forum Advisory Note
This forum is moderated by volunteer moderators who will react only to members' feedback on posts. Moderators are not employees or representatives of HWZ Forums. Forum members and moderators are responsible for their own posts. Please refer to our Community Guidelines and Standards and Terms and Conditions for more information.
Top