Loading...
> >

Remixing your art with AI

1…161718192022»

Comments

  • ArtiniArtini Posts: 9,508
    edited December 10

    The important is that it brings you so much fun, @WendyLuvsCatz

    I am still have not enough spare time to play with animations.

    Great, that we have a FilaToon now, so I can test some ideas in Daz Studio.

     

    Post edited by Artini on
  • WendyLuvsCatzWendyLuvsCatz Posts: 38,302

    Ha, I learned there is a random user with zero posts simply called Wendy

    just add an @ to find them all

    @DAZ 

     

  • DiomedeDiomede Posts: 15,184
    edited December 10

    Stopping by to see if anyone had comments on the new releases by OpenAI, including but not limited to SORA.  

    Post edited by Diomede on
  • WendyLuvsCatzWendyLuvsCatz Posts: 38,302

    https://www.genmo.ai/play

    not openAI but a new free Video generator, open Source too, hopefully builds for A1111, ComfyUI, Pinokio etc will come

  • wolf359wolf359 Posts: 3,829
    edited December 10

    Diomede said:

    Stopping by to see if anyone had comments on the new releases by OpenAI, including but not limited to SORA. 
     

    Many of the AI video enthusiasts on X(twitter)
    report that even the $200 per month new version of Sora is “underwhelming” and lagging behind it’s competitors even some of the free ones.

    Post edited by wolf359 on
  • ArtAngelArtAngel Posts: 1,738
    edited December 11

    Had to postt this here. Two things.

    1. Not a fan of AI and do not utilize it, but FYI  Amazon is now asking authors to indicate if AI was used in either the cover or manuscript. Wondering if it will later relate to low content.

    2. This is amazing, and I really respect how AI pulled this one off. Watched the trailer for Tom Hanks 'HERE Movie'  $50 million dollar Movie made with Generative AI to de-age him and holy crap, it blew me away! 

    Here's an interview about it.

    Post edited by ArtAngel on
  • WendyLuvsCatzWendyLuvsCatz Posts: 38,302
    edited December 11

    I have installed another local one in Pinokio called COG

    it does text to video (pretty bad, Modelscope is better)

    image to video (not bad, better than some online ones but Kilng, Pika, Haiper, Luma and MiniMax still better, not tried Luma)

    it does take a while but actually doesn't use that much VRAM about 7GB

    the interpolation is from RIFE because it's only 8fps hence the blurring, does 6 seconds of video though

    upscales to 4K too, I added lipsync and faceswap (face was bit blurry and made it higher resolution) in FaceFusion to this, it's also a Fooocus edited image to make the Impressionist Manet painting more photorealistic 

    COG can extend videos too and change the subject of video with prompts but not tried that yet as will involve hours of more model downloading and my internet rather slow lately, worse I had 5 power outages one for 7 hours the past 2 days

    Post edited by WendyLuvsCatz on
  • WendyLuvsCatzWendyLuvsCatz Posts: 38,302

    this might be of interest to many as trained on Public Domain content 

    I will certainly look into using this model if someone can create a webui that's not node based like ComfyUI

    hopefully a checkpoint for A1111 or Fooocus

  • DiomedeDiomede Posts: 15,184

    Cool.  Thanks for the link.

    And, any thoughts on Pika v2?

    Forget Sora —Pika Labs drops v2 of its creative AI video model

  • WendyLuvsCatzWendyLuvsCatz Posts: 38,302

    Diomede said:

    Cool.  Thanks for the link.

    And, any thoughts on Pika v2?

    Forget Sora —Pika Labs drops v2 of its creative AI video model

    no longer available to me without a subscription

    I did get a lot of use out of the old verson so I guess that was inevitable 

Sign In or Register to comment.