Preprint
Aug 2026
Dynamic-Robust Photometric-Semantic Reconstruction for Open-Vocabulary 3D Scene Understanding
SPAR is proposed, a novel joint semantic-geometric encoding architecture that explicitly isolates transient dynamic noise prior to latent space aggregation that structurally couples motion estimation with multi-view visual and semantic learning and reveals a strong inter-task synergy between photometric scene reconstru...
Boyu Cai, Li Yang, Yan Xu et al.
· 0 citations