Selecting a complete 3D object from a reconstructed scene with minimal user effort is essential for practical scene editing and embodied interaction. Existing 3D Gaussian Splatting (3DGS) methods either retrain the Gaussian representation to embed per-object labels or build dense multi-view SAM observations, both requiring heavy computation and dense viewpoint coverage that is rarely available in practice.
GaussianSelector, presented by Hugging Face researchers, is a training-free framework for interactive 3D object selection from sparse views and sparse scribble guidance. It operates directly on native Gaussian primitives, coarsening dense Gaussians into geometrically coherent superpoints and constructing a continuity-weighted graph using appearance and spatial cues.
Sparse user scribbles are lifted into 3D via visibility-aware transmittance coverage, and selection is solved as a global graph-cut energy minimization that propagates sparse evidence to a complete 3D object. This design naturally supports multi-round refinement, where users iteratively correct the selection from additional viewpoints to progressively improve the result.
Experiments demonstrate that GaussianSelector achieves competitive selection quality against state-of-the-art multi-view SAM-based methods, while requiring significantly fewer interaction views and substantially lower computational overhead. These properties make it well suited for human-in-the-loop 3D scene editing and 3D asset extraction in real-world deployment scenarios.