CSMapping: Scalable Crowdsourced Semantic Mapping and Topology Inference for Autonomous Driving

TL;DR

CSMapping leverages a diffusion-based generative prior and robust optimization techniques to produce high-quality, scalable semantic and topological maps for autonomous driving from crowdsourced data, despite sensor noise.

Contribution

It introduces a novel diffusion model-based approach for semantic map generation and a robust clustering method for topological mapping, improving quality with more data.

Findings

Achieves state-of-the-art semantic mapping accuracy.

Produces smooth, human-like topological road centerlines.

Demonstrates scalability and robustness on multiple datasets.

Abstract

Crowdsourcing enables scalable autonomous driving map construction, but low-cost sensor noise hinders quality from improving with data volume. We propose CSMapping, a system that produces accurate semantic maps and topological road centerlines whose quality consistently increases with more crowdsourced data. For semantic mapping, we train a latent diffusion model on HD maps (optionally conditioned on SD maps) to learn a generative prior of real-world map structure, without requiring paired crowdsourced/HD-map supervision. This prior is incorporated via constrained MAP optimization in latent space, ensuring robustness to severe noise and plausible completion in unobserved areas. Initialization uses a robust vectorized mapping module followed by diffusion inversion; optimization employs efficient Gaussian-basis reparameterization, projected gradient descent zobracket multi-start, and latent-space factor-graph for global consistency. For topological mapping, we apply confidence-weighted k-medoids clustering and kinematic refinement to trajectories, yielding smooth, human-like centerlines robust to trajectory variation. Experiments on nuScenes, Argoverse 2, and a large proprietary dataset achieve state-of-the-art semantic and topological mapping performance, with thorough ablation and scalability studies.