GSMap: 2D Gaussians for Online HD Mapping

TL;DR

GSMap introduces a unified 2D Gaussian framework for online HD mapping, effectively combining geometric precision and topological structure to improve autonomous driving map construction.

Contribution

It proposes a novel learnable Gaussian representation that unifies rasterization and vectorization approaches for HD map creation.

Findings

Achieves significant performance improvements on nuScenes and Argoverse2 datasets.

Effectively combines geometric fidelity with topological regularity.

Demonstrates compatibility with existing HD mapping architectures.

Abstract

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