Text2LIVE: Text-Driven Layered Image and Video Editing

Have you guys checked this out? Starting with an image (or sequence of images) plus a text label, the NN modifies consistently the given image to include the label.

Here is a compilation of examples.

https://text2live.github.io/sm/index.html

I like video editing layers on images. I am using latest modded version of capcut, which is a video editing app. You can also give it a try here,