Problem
Research questions and friction points this paper is trying to address.
Next-Best-View Selection
3D Gaussian Splatting
Active 3D Reconstruction
Radiance Fields
AGILE-GS: Anchor-Guided Fast Next-Best-View Selection for Active 3D Gaussian Splatting
🤖 AI Summary
This study addresses the high computational cost and latency associated with next-best-view (NBV) selection in 3D Gaussian Splatting by proposing an anchor-guided NBV method. The core innovation lies in decoupling information search from camera selection: virtual anchors are optimized via Riemannian gradient ascent on the SE(3) manifold to maximize expected information gain, efficiently generating a non-redundant candidate pool without rendering; subsequently, a greedy ridge leverage algorithm performs rapid screening. Experiments demonstrate that the proposed approach matches or surpasses existing baselines on standard benchmarks and closed-loop acquisition tasks, while reducing view selection latency by one to two orders of magnitude.
📝 Abstract
Radiance fields need hundreds of views, and their placement matters as much as their number. Next-best-view (NBV) selection for 3D Gaussian Splatting (3DGS) usually scores every candidate in the pool and keeps one. Searching for information and choosing a camera, however, are separable problems. We present AGILE-GS, an anchor-guided NBV method that separates the two. A virtual anchor pose is optimized on SE(3) by Riemannian gradient ascent on expected information gain. It need not be reachable or in the pool; it marks where the model is most uncertain. Candidates are scored against the anchor's viewing geometry, and a greedy ridge-leverage step distills the pool into a small, non-redundant shortlist without rendering any candidate. The shortlist can be used in two ways. AGILE-GS takes the first view on it as the next view, so no Fisher information is computed for any candidate. AGILE-GS+ computes the Fisher information gain of each shortlisted view and picks the best, so the expensive evaluation runs on a handful of views rather than the whole pool. On standard benchmarks and in closed-loop embodied acquisition, both match or exceed existing baselines while cutting selection latency by one to two orders of magnitude.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Next-Best-View Selection
3D Gaussian Splatting
Riemannian Gradient Ascent
Anchor-Guided Optimization
Active Vision
🔎 Similar Papers
No similar papers found.
