用可学习的2D高斯点统一地图几何与拓扑，提升自动驾驶高清地图精度。
GSMap: 2D Gaussians for Online HD Mapping
- 用有序2D高斯序列建模地图元素，对应向量化的顶点位置。
- 在nuScenes和Argoverse2上实现几何与拓扑联合优化，性能显著提升。
- 兼容现有高清地图架构，适合自动驾驶场景下的实时地图构建。
高精地图构建对自动驾驶至关重要，但现有方法存在根本性权衡：基于矢量的方法保持拓扑结构但几何保真度差，基于栅格的方法能实现精确几何监督但输出无结构。为此，我们提出GSMap，一种通过可学习的2D高斯表示统一两种范式的新型框架。每个地图元素被建模为有序的2D高斯序列，其中心对应矢量化折线/多边形的顶点。该形式支持同时优化：(1) 可微栅格化以施加像素级几何约束；(2) 拓扑感知矢量化以维持结构规则性。在nuScenes和Argoverse2上的实验表明，基于高斯的表示有效融合几何与拓扑学习，在性能上实现显著提升，并展现出与现有高清地图架构的良好兼容性。代码将发布于https://github.com/peakpang/GSMap。
原文摘要 · Abstract (English)
Accurate High-Definition (HD) map construction is critical for autonomous driving, yet existing methods face a fundamental trade-off: vectorization-based approaches preserve topology but struggle with geometric fidelity, while rasterization-based approaches enable precise geometric supervision but produce unstructured outputs. To bridge this gap, we propose GSMap, a novel framework that unifies both paradigms via a learnable 2D Gaussian representation. Each map element is modeled as an ordered sequence of 2D Gaussians, whose centers correspond to the vertices of the vectorized polyline/polygon. This formulation enables simultaneous optimization through: (1) Differentiable rasterization that enforces pixel-level geometric constraints, and (2) Topology-aware vectorization that maintains structural regularity. Experiments on both nuScenes and Argoverse2 demonstrate that our Gaussian-based representation effectively unifies geometric and topological learning, achieving significant performance improvements and demonstrating strong compatibility with existing HD mapping architectures. Code will be available at https://github.com/peakpang/GSMap
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。