Interpolate between two images (FILM - Hugging Face Spaces Tutorial)

Поделиться
HTML-код
  • Опубликовано: 7 сен 2024
  • Join this channel to get access to perks:
    / @artificialimages
    Learn more about machine learning for image makers by signing up at mailchi.mp/da9...
    Join the Artificial Images Slack channel: join.slack.com...
    artificial-imag...
    / bustbright
    bustbright.com/
    / dvsmethid
    / dvsch
    dvschultz.gith...

Комментарии • 14

  • @Endgame10156
    @Endgame10156 Год назад +9

    IT'S FUCKING BROKEN ._.

  • @raioxnoticias
    @raioxnoticias 2 года назад +6

    how to create an interpolation video with more than 2 images ?

    • @yeahbutcanyouredacted3417
      @yeahbutcanyouredacted3417 10 месяцев назад

      RunwayML has a free frame Interpolation you can upload a whole batch of img's to interpolate through- im here to find if there is a way to do so on my own PC lol

  • @officialSgtPepperArc360
    @officialSgtPepperArc360 9 месяцев назад +2

    *Runtime error*

  • @elbedouirayen7938
    @elbedouirayen7938 Год назад +2

    cool can we use more than two frames please??
    thank u

  • @LexTrickett
    @LexTrickett 2 года назад +2

    Thank you so much for sharing. How would i go about running this locally so the export size is larger?

    • @ArtificialImages
      @ArtificialImages  2 года назад +1

      if you need higher resolution outputs you’ll need to look at the FILM repo to break videos into blocks. I doubt local will help with that unless you have a massive GPU locally

  • @hatoviah5790
    @hatoviah5790 2 года назад +3

    Thanks for the videos, would love to see this in the colab notebook.
    Quick questions: Is there a site where I can find a list of different colab notebooks to try out and a site that provides text prompts for various image output styles?
    Thank you

    • @eatapeach993
      @eatapeach993 2 года назад +4

      AI Notebooks and other sites:
      docs.google.com/document/d/1ON4unvrGC2fSEAHMVb4idopPlWmzM0Lx5cxiOXG47k4/edit
      200+ CLIP+VQGAN examples with 4 different subjects:
      imgur.com/a/SALxbQm

    • @hatoviah5790
      @hatoviah5790 2 года назад +1

      @@eatapeach993 Extremely helpful resources. Thank you :)

  • @sluggy6074
    @sluggy6074 11 месяцев назад

    Can you use videos with alpha channels? That would probably help. Itd be better to have that than train the AI to focus on the people or the features cause thatd likely hamper its ability to smooth out environmental videos when you need them like a river or something.

  • @drekenproductions
    @drekenproductions Год назад

    is there censoring? if i add these two specific images with some cartoon skeletons the images return black. ive tried different file formats aspect ratios everything. it rejects the images and gives black output. even tried taking a screenshot with gyazo and downloading fresh image files. the outputs always black. test images work. images i dont care about work but not the cartoon skeletons. seems like a censorship issue like seeing the green screens or blurring

  • @seanzhang3873
    @seanzhang3873 5 месяцев назад

    yo, it's broken