Monocular 3D Lane Detection via Structure Uncertainty-Aware Network with Curve-Point Queries

TL;DR

MonoUnc introduces a novel monocular 3D lane detection method that explicitly models structural uncertainty using curve-based representations and Gaussian modeling, improving accuracy over previous approaches.

Contribution

The paper presents MonoUnc, a BEV-free 3D lane detection framework that explicitly incorporates local structural and aleatoric uncertainties through curve-point queries and Gaussian modeling.

Findings

Outperforms state-of-the-art methods on ONCE-3DLanes and OpenLane datasets.

Introduces new evaluation metrics based on bidirectional Chamfer distances.

Effectively models local lane structure and uncertainty for improved detection accuracy.

Abstract

Monocular 3D lane detection is challenged by aleatoric uncertainty arising from inherent observation noise. Existing methods rely on simplified geometric assumptions, such as independent point predictions or global planar modeling, failing to capture structural variations and aleatoric uncertainty in real-world scenarios. In this paper, we propose MonoUnc, a bird's-eye view (BEV)-free 3D lane detector that explicitly models aleatoric uncertainty informed by local lane structures. Specifically, 3D lanes are projected onto the front-view (FV) space and approximated by parametric curves. Guided by curve predictions, curve-point query embeddings are dynamically generated for lane point predictions in 3D space. Each segment formed by two adjacent points is modeled as a 3D Gaussian, parameterized by the local structure and uncertainty estimations. Accordingly, a novel 3D Gaussian matching loss is designed to constrain these parameters jointly. Experiments on the ONCE-3DLanes and OpenLane datasets demonstrate that MonoUnc outperforms previous state-of-the-art (SoTA) methods across all benchmarks under stricter evaluation criteria. Additionally, we propose two comprehensive evaluation metrics for ONCE-3DLanes, calculating the average and maximum bidirectional Chamfer distances to quantify global and local errors. Codes are released at https://github.com/lrx02/MonoUnc.