HGeo-TopoMap: Boosting Topological Mapping with Hierarchical Geometric Priors

2026-07-23

AI Summary

This work addresses the challenge of centerline instance detection in real-world scenarios where road markings are often missing or degraded. To tackle this problem, the authors propose a hierarchical topological map construction method that integrates explicit geometric priors with implicit spatial relationships. The approach first employs a geometry-adaptive learning module to extract semantic and spatial features from inverse perspective-mapped road structure images, enhanced by a prior mask attention mechanism to focus on salient regions. Subsequently, a geometry-consistency learning module aligns co-directional centerline features within a geometry-aware decoder to reinforce spatial coherence. Evaluated on the OpenLane-V2 dataset, the method significantly outperforms existing approaches in centerline detection, lane segment reconstruction, and robustness benchmarks, demonstrating superior accuracy and stability under both standard and challenging conditions.

Abstract

Topological maps are key outputs of autonomous driving perception systems, delivering essential road information for path planning. They identify instances such as centerlines and traffic signs, along with their connectivity relationships. Due to the lack of explicit markings for centerlines in real-world environments, the detection of centerline instances remains a significant challenge. To tackle this problem, we propose HGeo-TopoMap, which leverages an explicit prior map and implicit spatial relations to hierarchically boost topological mapping. First, a geometric adaptive learning module is designed for the road structure map obtained via inverse perspective mapping. This module discretely encodes semantic and spatial features from the map, followed by a prior-mask attention mechanism that selectively focuses on informative regions. Then, a geometric consistency learning module is devised, which leverages the geometric properties and spatial relationships of centerlines. Built on the geometry-aware decoder, it enforces spatial consistency by aligning features of centerline instances with identical geometric orientations. The proposed method is evaluated on the OpenLane-V2 dataset across the centerline, lane segment, and robustness benchmarks. Beyond substantial improvements in topological mapping accuracy, the proposed method offers the benefit of enhanced robustness, consistently outperforming baselines under both standard and challenging conditions. The source code and model weights will be made publicly available at https://github.com/lynn-yu/HGeo-TopoMap.

Problem

• topological mapping
• centerline detection
• autonomous driving
• geometric priors
• road perception

Innovation

• Hierarchical Geometric Priors
• Topological Mapping
• Geometric Consistency Learning
• Prior-Mask Attention
• Centerline Detection