We refer to the proposed solution as SA3D, for Segment Anything in 3D. It is only required to provide a manual segmentation prompt ( e.g., rough points) for the target object in a single view, which is used to generate its 2D mask in this view with SAM.

The ability to sense and reconstruct 3D appearance and geometry is critical to applications in vision, graphics, and beyond.

In this paper, we propose the M asked S pace-T ime H ash encoding (MSTH), a novel method for efficiently reconstructing dynamic 3D scenes from multi-view or monoc-ular videos.

These estimates are subsequently utilized to aggregate information from reference frames into the target frame. Subsequently, hierarchical neural point clouds are constructed based on the aggregated information. This hierarchical point cloud set is then employed to synthesize views of the target frame.

It represents a radiance field using tri-planes with three multi-resolutions for each plane: 128, 256, and 512 in both height and width, and 32 in feature depth. However, any MDE model can be utilized within our framework [19, 13, 12]. The training process takes approximately 3 hours. In other words, we can rewrite the above scheme as a closed problem. The results of DDP-NeRF with in-domain priors are 20.96,

Project page is available at https://ku-cvlab.github.io/DaRF/ .

Several recent attempts have been made on neural reconstruction from sparse views.