VS-Splat: Voxel-Selective feed-forward Gaussian Splatting for end-to-end 3D object reconstruction from sparse-views

arXiv:2609.12343v1 Announce Type: new Abstract: Feed-forward Gaussian splatting models have demonstrated remarkable effectiveness in reconstructing three-dimensional (3D) objects from a few two-dimensional (2D) images, even if they are unseen. As existing methods typically predict Gaussian primitives uniformly across the 3D space, most primitives are placed in non-object regions. This may hinder the representation of fine object details. This paper proposes a Voxel-Selective Gaussian Splatting model (VS-Splat), a new end-to-endfeed-forward Gaussian splatting framework that predicts many primitives only within selected voxels that are likely to belong to an object, without 3D structural supervision. To achieve this, we propose a new learnable voxel selection approach that identifies object-centric voxels only with 2D rendering supervision.Our sparse-view rendering experiments with three benchmark datasets show that proposed VS-Splat outperforms several state-of-the-art methods. We further demonstrate its effectiveness as a backbone for an existing densification method and show that anoptional extension improves its robustness to inaccurate camera pose estimates.

Related

• DelowlightSplat: Feed-Forward Gaussian Splatting for Lowlight 3D Scene Reconstruction
• SPAGS: Sparse-View Articulated Object Reconstruction from Single State via Planar Gaussian Splatting
• SkySplat: Generalizable 3D Gaussian Splatting from Multi-Temporal Sparse Satellite Images
• AnchorSplat: Feed-Forward 3D Gaussian Splatting with 3D Geometric Priors
• ArtSplat: Feed-Forward Articulated 3D Gaussian Splatting from Sparse Multi-State Uncalibrated Views

Source: arXiv cs.CV | 2026-09-14