Home | Publications | WLL+26

Visibility-Aware Language Aggregation for Open-Vocabulary Segmentation in 3D Gaussian Splatting

MCML Authors

Abstract

Recently, distilling open-vocabulary language features from 2D images into 3D Gaussians has attracted significant attention. Although existing methods achieve impressive language-based interactions with 3D scenes, we observe two fundamental issues: background Gaussians, which contribute negligibly to a rendered pixel, receive the same feature as the dominant foreground ones, and multi-view inconsistencies due to view-specific noise in language embeddings. We introduce Visibility-Aware Language Aggregation (VALA), a lightweight yet effective method that computes marginal contributions for each ray and applies a visibility-aware gate to retain only visible Gaussians. Moreover, we propose a streaming weighted geometric median in cosine space to merge noisy multi-view features. Our method yields a robust, view-consistent language feature embedding in a fast and memory-efficient manner. VALA improves open-vocabulary localization and segmentation across reference datasets, consistently surpassing existing works.

inproceedings WLL+26

3DV 2026

13th International Conference on 3D Vision. Vancouver, Canada, Mar 20-23, 2026. To be published. Preprint available.

Authors

S. Wang • K. Li • S. Liang • E. Alegret • J. Ma • N. Navab • S. Gasperini

Links

URL GitHub

In Collaboration

 VisualAIs Labs

Research Area

 C1 | Medicine

BibTeXKey: WLL+26

#p-navab #p-tombari
Back to Top