Unified Modeling of Lane and Lane Topology for Driving Scene Reasoning

2026-05-09

AI Summary

This work addresses the limitations of conventional “detect-then-reason” paradigms in autonomous driving, which struggle to efficiently model lane geometries and their topological relationships. To overcome this, the authors propose UniTopo, an end-to-end unified perception framework that jointly encodes lane geometry and topological relations—such as predecessor, successor, and intersection—as a connected structure, enabling simultaneous prediction of lane positions and connectivity within a shared feature space. By eliminating the traditional two-stage pipeline, UniTopo achieves direct, topology-aware visual reasoning. Evaluated on the OpenLane-V2 benchmark, UniTopo attains TOP_ll scores of 30.1% and 31.8% on two subsets, outperforming the current state-of-the-art method T²SG by 6.0% and 8.6%, respectively.

Abstract

Autonomous vehicles need to perceive not only physical elements in the driving scene, such as lane lines and traffic lights, but also logical elements like lane centerlines and their topology. Existing lane topology reasoning methods typically follow a reasoning-by-detection paradigm, where lane topological relationships are primarily derived from lane detection results. In this paper, we propose an innovative method called Unified Modeling of Lane and Lane Topology (UniTopo), which represents the topological relationships between lanes as connected lanes, encompassing predecessor lanes, successor lanes, and their interconnections. This unified representation of lanes and lane topology allows us to simultaneously obtain both the positions and topological information of lanes within a shared perception pipeline, establishing a new paradigm for directly perceiving lane topology from original image features. We validate our method on the driving scene reasoning benchmark OpenLane-V2, which consists of two subsets, built based on Argoverse2 and nuScenes, respectively. Our method achieves TOP_ll of 30.1% and 31.8% on the two subsets, significantly surpassing the existing state-of-the-art method T^2SG by 6.0% and 8.6%.

Problem

• lane topology
• driving scene reasoning
• autonomous driving
• lane perception
• topological relationships

Innovation

• lane topology
• unified modeling
• driving scene reasoning
• connected lanes
• perception pipeline