Human-centric videos demand higher facial clarity. We added difix restoration and cross-frame static spatial supervision reuse to our pipeline — sharper faces, more stable backgrounds.
Single video → a reframeable 4D Gaussian Splatting scene.
Not a sequence of separately built 3D frames played back like a video. This is one continuous space-time scene, reconstructed from a single clip shot on an iPhone 16.
We combine feed-forward Gaussian generation, 3D
