NVIDIA Text2LIVE modifies semantic regions in photos & vids

When it rains, it pours: Text2LIVE promises the ability to use descriptions to modify parts of photos:

[O]ur goal is to edit the appearance of existing objects (e.g., object’s texture) or augment the scene with new visual effects (e.g., smoke, fire) in a semantically meaningful manner.

It also works on video:

Leave a Reply

Your email address will not be published. Required fields are marked *