Our method achieves fast⚡️ and robust⛷️ dynamic-static decomposition based on 3D/4D gaussian splatting for a wide range of inputs as long egocentric videos, image collections, multi-view/monocular ...
While it's great to have this show back on our screens, it still doesn't soothe the sting of its cancellation in the first ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results