Generate dynamic 3D gaussians from multiple videos?

[adrida]

Hello, thank you for the great work. I was wondering if it was possible with the current implementation to pass as input a few videos from a new scene and get a 3d animation file that could be rendered using Blender for example? I was thinking of something little bit like the Nerfs approaches, in a way it would be like generating novel views from the ones in the video. Or do we necessarily need a 3d asset as input?