MAGICIAN: Efficient Long-Term Planning with Imagined Gaussians for Active Mapping

TL;DR

MAGICIAN introduces a long-term planning framework for active mapping that leverages Imagined Gaussians for efficient environment reconstruction, outperforming greedy methods in indoor and outdoor benchmarks.

Contribution

The paper presents MAGICIAN, a novel long-term planning approach using Imagined Gaussians for efficient exploration and mapping, integrating scene representation with tree-search planning.

Findings

Achieves state-of-the-art results on multiple benchmarks.

Effectively balances exploration and reconstruction quality.

Outperforms greedy next-best-view methods.

Abstract

Active mapping aims to determine how an agent should move to efficiently reconstruct unknown environments. Most existing approaches rely on greedy next-best-view prediction, resulting in inefficient exploration and incomplete reconstruction. To address this, we introduce MAGICIAN, a novel long-term planning framework that maximizes accumulated surface coverage gain through Imagined Gaussians, a scene representation based on 3D Gaussian Splatting, derived from a pre-trained occupancy network with strong structural priors. This representation enables efficient coverage gain computation for any novel viewpoint via fast volumetric rendering, allowing its integration into a tree-search algorithm for long-horizon planning. We update Imagined Gaussians and refine the trajectory in a closed loop. Our method achieves state-of-the-art performance across indoor and outdoor benchmarks with varying action spaces, highlighting the advantage of long-term planning in active mapping.