MGMap: Mask-Guided Learning for Online Vectorized HD Map Construction

TL;DR

MGMap introduces a mask-guided learning framework for online HD map construction, enhancing map element localization and structural detail preservation through learned masks and multi-scale BEV features.

Contribution

The paper proposes MGMap, a novel mask-guided approach with MAI decoder and PG-MPR module for improved accuracy and robustness in online vectorized HD map construction.

Findings

Achieves around 10 mAP improvement over baselines

Demonstrates strong robustness and generalization

Effective in highlighting informative regions for map elements

Abstract

Currently, high-definition (HD) map construction leans towards a lightweight online generation tendency, which aims to preserve timely and reliable road scene information. However, map elements contain strong shape priors. Subtle and sparse annotations make current detection-based frameworks ambiguous in locating relevant feature scopes and cause the loss of detailed structures in prediction. To alleviate these problems, we propose MGMap, a mask-guided approach that effectively highlights the informative regions and achieves precise map element localization by introducing the learned masks. Specifically, MGMap employs learned masks based on the enhanced multi-scale BEV features from two perspectives. At the instance level, we propose the Mask-activated instance (MAI) decoder, which incorporates global instance and structural information into instance queries by the activation of instance masks. At the point level, a novel position-guided mask patch refinement (PG-MPR) module is designed to refine point locations from a finer-grained perspective, enabling the extraction of point-specific patch information. Compared to the baselines, our proposed MGMap achieves a notable improvement of around 10 mAP for different input modalities. Extensive experiments also demonstrate that our approach showcases strong robustness and generalization capabilities. Our code can be found at https://github.com/xiaolul2/MGMap.