Summary
GAVIS (Gaussian Splatting Anisotropic Visibility Field) is a new framework for uncertainty quantification and active mapping in 3D Gaussian Splatting (3DGS). Its core insight is that regions unseen by training views yield unreliable predictions in 3DGS. The authors introduce a principled and efficient method to quantify the visibility field in 3DGS, defined as the anisotropic visibility of each particle relative to training views and represented with spherical harmonics. This visibility field is integrated into a Bayesian-network-based uncertainty-aware 3DGS rasterizer, enabling real-time uncertainty quantification of synthesized views at 200 FPS. Active mapping is then achieved through a maximum information gain framework. Extensive experiments across diverse environments show GAVIS consistently and significantly outperforms prior methods in accuracy and efficiency. Additionally, beyond standalone use, the method can be applied post hoc to improve the performance of existing approaches. Paper: arXiv 2605.30342, by Shangjie Xue, Jesse Dill, Dhruv Ahuja, Frank Dellaert, Panagiotis Tsiotras, and Danfei Xu.
Paper Overview
Field: Computer Vision
Authors: Shangjie Xue, Jesse Dill, Dhruv Ahuja, Frank Dellaert, Panagiotis Tsiotras, Danfei Xu
Published: 2026-05-28
arXiv: 2605.30342
Summary
We present GAVIS (Gaussian Splatting Anisotropic Visibility Field), a novel framework for uncertainty quantification and active mapping in 3D Gaussian Splatting (3DGS). The core insight is that regions not observed by training views produce unreliable predictions in 3DGS. To address this, we introduce a principled and efficient approach for quantifying the visibility field in 3DGS, defined as the anisotropic visibility of each particle with respect to the training views, represented via spherical harmonics.
The resulting visibility field is integrated into a Bayesian-network-based uncertainty-aware 3DGS rasterizer, enabling real-time uncertainty quantification (200 FPS) of synthesized views. Active mapping is then achieved through a maximum information gain framework.
Extensive experiments across diverse environments demonstrate that GAVIS consistently and significantly outperforms prior methods in both accuracy and efficiency. Moreover, beyond standalone use, our method can be applied post hoc to improve the performance of existing approaches.
---
*Auto-collected on 2026-06-01*
This page is an English static mirror generated for search and AI citation.
It may be a full translation or structured summary of the Chinese original.
Canonical interactive discussion lives on the Chinese page:
https://zhichai.net/topic/177980676