Peking University, HKUST, Pengcheng Laboratory · 3D Reconstruction

PURE

An ACM Multimedia 2026 paper that post-trains feed-forward 3D reconstruction models on unlabeled video to improve them on dynamic scenes, with no ground-truth 3D annotations. Training cues come from monocular depth, dynamic masks and cross-frame matches, covering depth ordering, local geometry and static consistency. It is applied to three backbones, Pi3, Depth Anything 3 and VGGT, and outputs depth, camera poses and colored point clouds from video or image sequences. The project shows a 2,000-frame reconstruction and offers code on GitHub, weights on Hugging Face and a PURE Studio demo Space.