# L2G-Map: Local-to-Global Mapping via Hierarchical Diffusion Refinement and Elliptical Bayesian Fusion

arXiv: https://arxiv.org/abs/2503.02578 · HTML: https://mlsanity.com/paper/2503.02578/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 30 references that are in the map.
- First posted on arXiv 4 Mar 2025.

## Builds on (most cited first, 19 of 19)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Spatial Transformer Networks](https://mlsanity.com/paper/1506.02025/) (2015)
- [Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D](https://mlsanity.com/paper/2008.05711/) (2020)
- [Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting](https://mlsanity.com/paper/2301.00493/) (2023)
- [NVAE: A Deep Hierarchical Variational Autoencoder](https://mlsanity.com/paper/2007.03898/) (2020)
- [HDMapNet: An Online HD Map Construction and Evaluation Framework](https://mlsanity.com/paper/2107.06307/) (2021)
- [DDP: Diffusion Model for Dense Visual Prediction](https://mlsanity.com/paper/2303.17559/) (2023)
- [StreamMapNet: Streaming Mapping Network for Vectorized Online HD Map Construction](https://mlsanity.com/paper/2308.12570/) (2023)
- [Neural Map Prior for Autonomous Driving](https://mlsanity.com/paper/2304.08481/) (2023)
- [P-MapNet: Far-seeing Map Generator Enhanced by both SDMap and HDMap Priors](https://mlsanity.com/paper/2403.10521/) (2024)
- [MapPrior: Bird's-Eye View Map Layout Estimation with Generative Models](https://mlsanity.com/paper/2308.12963/) (2023)
- [DiffMap: Enhancing Map Segmentation with Map Prior Using Diffusion Model](https://mlsanity.com/paper/2405.02008/) (2024)
- [GlobalMapNet: An Online Framework for Vectorized Global HD Map Construction](https://mlsanity.com/paper/2409.10063/) (2024)
- [BLOS-BEV: Navigation Map Enhanced Lane Segmentation Network, Beyond Line of Sight](https://mlsanity.com/paper/2407.08526/) (2024)
- [DTCLMapper: Dual Temporal Consistent Learning for Vectorized HD Map Construction](https://mlsanity.com/paper/2405.05518/) (2024)

## Abstract (arXiv)

Offline high-definition maps provide essential geometric and topological priors for autonomous driving systems. Pure-vision solutions have become the predominant paradigm for offline mapping due to their cost-effectiveness and scalability. However, local-to-global mapping under visual conditions confronts two fundamental challenges: single-shot local observations are susceptible to viewpoint variation and environmental interference, leading to geometric deviations, while multi-source local information exhibits heterogeneous confidence, rendering globally consistent aggregation difficult. To address these, this paper proposes L2G-Map, a framework comprising hierarchical prior diffusion refinement and elliptical space Bayesian fusion. The former jointly embeds temporal context and centerline priors to guide structure completion and topology recovery during denoising, alleviating the information incompleteness inherent in pure-vision settings. The latter incorporates an adaptive weighting strategy driven by elliptical distance propagation, enabling probabilistically optimal aggregation of multi-source information under the Bayesian posterior update paradigm. Extensive experiments on nuScenes and Argoverse benchmark datasets verify the effectiveness of L2G-Map. The proposed refinement component yields consistent local map accuracy improvements across different datasets. Under sensor-degraded conditions, a 3.27% mIoU gain is achieved. Furthermore, the adaptive fusion component significantly enhances the accuracy of global maps. The fused global map can be flexibly embedded into different online map models, yielding an 18.26% mIoU improvement in semantic map construction and a 20.00% enhancement in vectorized map construction, demonstrating the overall advantages of the proposed closed-loop pipeline. Source code will be available at https://github.com/lynn-yu/L2G-Map.