Jul 2026
MonoVoc: Decoupling Geometry and Semantics for Lightweight Monocular Open-Vocabulary 3D Gaussians
This work presents a novel, training-free pipeline that fundamentally reimagines this paradigm by explicitly decoupling 3D geometric reconstruction from semantic integration, and delivers an order-of-magnitude reduction in memory usage compared to SOTA baselines.
Pouya Ardekhani, Zahra Dehghanian, Morteza Abolghasemi et al.
· arXiv.org · 0 citations