MAGICIAN：利用想象高斯进行高效长时程规划以完成主动建图
MAGICIAN: Efficient Long-Term Planning with Imagined Gaussians for Active Mapping
把预训练占用网络转成可快速体渲染的「想象高斯」，从而在树搜索里低成本评估任意新视角的表面覆盖增益——这套「用可微渲染代理函数做长时程规划打分」的思路可直接迁移到主动重建、下一最佳视角选择和具身探索策略学习。
原文依据 maximizes accumulated surface coverage gain through Imagined Gaussians, a scene representation derived from a pre-trained occupancy network
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论主动建图旨在确定智能体应如何移动，以高效重建未知环境。现有方法大多依赖贪心的下一最佳视角预测，导致探索效率低下、场景重建不完整。为解决这一局限，我们提出 MAGICIAN，一种新颖的长时程规划框架，它通过「想象高斯」最大化累积表面覆盖增益；该场景表示由具有强结构先验的预训练占用网络导出。这一表示借助快速体渲染，能够高效计算任意新视角的覆盖增益，从而可将其集成到树搜索算法中以实现长时程规划。我们以闭环方式更新想象高斯并细化规划轨迹。我们的方法在具有不同动作空间的室内与室外基准上取得了最优性能，凸显了长时程规划在主动建图中的关键优势。
英文原文摘要 996 字符
Active mapping aims to determine how an agent should move to efficiently reconstruct an unknown environment. Most existing approaches rely on greedy next-best-view prediction, resulting in inefficient exploration and incomplete scene reconstruction. To address this limitation, we introduce MAGICIAN, a novel long-term planning framework that maximizes accumulated surface coverage gain through Imagined Gaussians, a scene representation derived from a pre-trained occupancy network with strong structural priors. This representation enables efficient computation of coverage gain 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 planned trajectory in a closed-loop manner. Our method achieves state-of-the-art performance across indoor and outdoor benchmarks with varying action spaces, demonstrating the critical advantage of long-term planning in active mapping.