MotionGS-SLAM：面向运动模糊鲁棒SLAM的事件调制高斯溅射
MotionGS-SLAM: Event-Modulated Gaussian Splatting for Motion-Blur Robust SLAM
浏览论文内容
中文总结 AI 辅助
MotionGS-SLAM通过事件调制高斯核与双调制机制，在渲染流水线中建模运动模糊形成，实现相机轨迹与场景几何联合优化，显著提升高运动下SLAM的轨迹与地图精度。
中文摘要 AI 辅助
当前基于视觉的SLAM系统在运动模糊破坏视觉输入时会出现灾难性失效，因为它们试图解决从退化观测中恢复清晰内容这一不适定逆问题。我们提出MotionGS-SLAM，该方法通过范式转变从根本上重新构想运动模糊处理：我们不消除模糊伪影，而是将这一挑战重新表述为一个约束良好的正问题，在渲染流水线中生成式地建模模糊形成过程。利用事件相机的微秒级时间分辨率和抗运动模糊特性，我们引入了一种新颖的事件调制高斯核，可基于精确运动线索动态调整每个高斯的光栅化。我们的双调制机制将2D高斯投影从各向同性点转换为各向异性、运动对齐的椭圆笔触（空间调制），同时基于局部速度自适应改变曝光积分采样密度（时间调制）。这种基于物理的方法通过感知模糊的光度约束和基于事件的约束，实现曝光内相机轨迹与3D场景几何的联合优化。大量实验表明，在剧烈高运动条件下，该方法在轨迹精度和地图质量方面相比最先进方法有显著提升。
英文摘要
Current Vision-based SLAM systems fail catastrophically when motion blur corrupts the visual input, as they attempt the ill-posed inverse problem of recovering sharp content from degraded observations. We present MotionGS-SLAM, which fundamentally reimagines motion blur handling through a paradigm shift: rather than removing blur artifacts, we reformulate the challenge as a well-constrained forward problem that generatively models blur formation within the rendering pipeline. By leveraging event cameras' microsecond temporal resolution and immunity to motion blur, we introduce a novel event-modulated Gaussian kernel that dynamically adapts each Gaussian's rasterization based on precise motion cues. Our dual-modulation mechanism transforms 2D Gaussian projections from isotropic dots into anisotropic, motion-aligned elliptical brush strokes (spatial modulation) while adaptively varying exposure integral sampling density based on local velocity (temporal modulation). This physics-based approach enables joint optimization of intra-exposure camera trajectories and 3D scene geometry through blur-aware photometric and event-based constraints. Extensive experiments demonstrate significant improvements over state-of-the-art methods in trajectory accuracy and map quality under severe high-motion conditions.
发表机构
- Research Institute for Science & Technology, Tokyo University of Science（东京理科大学科学技术研究所）
机构由 AI 辅助整理，请以论文原文为准。
