让SLAM在动态场景中稳定运行，靠的是像素级不确定性估计。
DROID-SLAM in the Wild
- 通过多视角特征不一致计算每像素不确定性
- 实测在复杂动态场景中达到顶尖定位与建图精度
- 适合需要实时鲁棒感知的机器人、AR应用
我们提出一种鲁棒的实时RGB SLAM系统，通过可微分的不确定性感知捆绑调整（Uncertainty-aware Bundle Adjustment）处理动态环境。传统SLAM假设场景静态，遇运动物体易失效。现有动态SLAM方法依赖预设动态先验或不确定性映射，但在未知动态物体或高度杂乱场景中仍受限。本文方法利用多视角视觉特征不一致性估计每像素不确定性，实现真实环境中稳定的跟踪与重建。系统在杂乱动态场景中达到当前最优相机位姿与场景几何精度，且以约10 FPS实时运行。代码与数据集见https://github.com/MoyangLi00/DROID-W.git。
原文摘要 · Abstract (English)
We present a robust, real-time RGB SLAM system that handles dynamic environments by leveraging differentiable Uncertainty-aware Bundle Adjustment. Traditional SLAM methods typically assume static scenes, leading to tracking failures in the presence of motion. Recent dynamic SLAM approaches attempt to address this challenge using predefined dynamic priors or uncertainty-aware mapping, but they remain limited when confronted with unknown dynamic objects or highly cluttered scenes where geometric mapping becomes unreliable. In contrast, our method estimates per-pixel uncertainty by exploiting multi-view visual feature inconsistency, enabling robust tracking and reconstruction even in real-world environments. The proposed system achieves state-of-the-art camera poses and scene geometry in cluttered dynamic scenarios while running in real time at around 10 FPS. Code and datasets are available at https://github.com/MoyangLi00/DROID-W.git.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。