search

lucidrains / deep-daze

Simple command line tool for text to image generation using OpenAI's CLIP and Siren (Implicit neural representation network). Technique was originally created by https://twitter.com/advadnoun
MIT License
4.37k stars 326 forks source link

Story does not generate video #113

Closed Fliens closed 3 years ago

Fliens commented 3 years ago

I was trying to recreate the example: image

But it did not generate the video. Is the video made in post or is deep-daze able to generate it?

It would also be nice if you could add the exact command the example was made with :)

NotNANtoN commented 3 years ago

Hi, sorry about the lack of information. I just quickly put that together.

The video is created in post by stitching all generated frames together with a command such as: ffmpeg -i "INPUT_TEXT.000%03d.jpg" -pix_fmt yuv420p movie.mp4

Unfortunately, I do not remember the exact command but something like this should work: imagine POEM_TEXT --num_layers 44 --batch_size 32 --create_story 1 --save_every 20 --image_width 256

Let me know if it works!