AI RESEARCH
UniSem: Generalizable Semantic 3D Reconstruction from Sparse Unposed Images
arXiv CS.CV
•
ArXi:2603.17519v1 Announce Type: new Semantic-aware 3D reconstruction from sparse, unposed images remains challenging for feed-forward 3D Gaussian Splatting (3DGS). Existing methods often predict an over-complete set of Gaussian primitives under sparse-view supervision, leading to unstable geometry and inferior depth quality. Meanwhile, they rely solely on 2D segmenter features for semantic lifting, which provides weak 3D-level and limited generalizable supervision, resulting in incomplete 3D semantics in novel scenes.