MOGS：在大场景中使用单目物体引导的高斯散射
MOGS: Monocular Object-guided Gaussian Splatting in Large Scenes
AI总结：
MOGS通过单目视觉-惯性传感器实现大场景中高斯散射的物体引导密集深度生成，显著降低训练时间和内存消耗，同时保持高质量渲染效果。
摘要：
近年来，3D高斯散射（3DGS）在生成逼真图像方面取得了显著进展，将其扩展到大场景中为自动驾驶等应用中的语义推理和预测开辟了新机会。目前，大场景中的最新系统主要源于基于激光雷达的管道，这些管道利用远距离深度感知。然而，它们需要昂贵的高通道传感器，其密集点云会消耗内存和计算资源，限制了可扩展性、车队部署和优化速度。我们提出了MOGS，一种单目3DGS框架，它用基于稀疏视觉-惯性（VI）结构从运动（SfM）线索的物体锚定、计量密集深度替代主动激光雷达深度。我们的核心思想是利用图像语义来假设每个物体的形状先验，将它们与稀疏但具有度量可靠性的SfM点锚定，并在每个物体上传播由此产生的度量约束，以生成密集深度。为解决两个关键挑战，即物体内部SfM覆盖不足和跨物体几何不一致，MOGS引入了（1）一个多尺度形状共识模块，它能够自适应地将小段合并到最符合SfM支持的粗略物体中，并用参数化形状模型拟合它们，以及（2）一个跨物体深度细化模块，该模块在结合几何一致性、先验锚定和边缘感知平滑性的组合目标下优化每个像素的深度。在公开数据集上的实验表明，使用低成本的VI传感器套件，MOGS将训练时间减少了高达30.4%，内存消耗减少了19.8%，同时在大场景中实现了高质量渲染，其效果与昂贵的基于激光雷达的方法相媲美。
英文摘要：
Recent advances in 3D Gaussian Splatting (3DGS) deliver striking photorealism, and extending it to large scenes opens new opportunities for semantic reasoning and prediction in applications such as autonomous driving. Today's state-of-the-art systems for large scenes primarily originate from LiDAR-based pipelines that utilize long-range depth sensing. However, they require costly high-channel sensors whose dense point clouds strain memory and computation, limiting scalability, fleet deployment, and optimization speed. We present MOGS, a monocular 3DGS framework that replaces active LiDAR depth with object-anchored, metrized dense depth derived from sparse visual-inertial (VI) structure-from-motion (SfM) cues. Our key idea is to exploit image semantics to hypothesize per-object shape priors, anchor them with sparse but metrically reliable SfM points, and propagate the resulting metric constraints across each object to produce dense depth. To address two key challenges, i.e., insufficient SfM coverage within objects and cross-object geometric inconsistency, MOGS introduces (1) a multi-scale shape consensus module that adaptively merges small segments into coarse objects best supported by SfM and fits them with parametric shape models, and (2) a cross-object depth refinement module that optimizes per-pixel depth under a combinatorial objective combining geometric consistency, prior anchoring, and edge-aware smoothness. Experiments on public datasets show that, with a low-cost VI sensor suite, MOGS reduces training time by up to 30.4% and memory consumption by 19.8%, while achieving high-quality rendering competitive with costly LiDAR-based approaches in large scenes.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
