HSDF-Lane: Height-Aligned Signed Distance Field with Semantic Lane Prior for 3D Lane Detection

2026-06-30

AI Summary

Monocular 3D lane detection is inherently challenged by depth ambiguity and road non-planarity, often leading to geometric distortions. This work proposes a novel approach that, for the first time, leverages signed distance fields (SDFs) to implicitly model the road surface, introducing a height-aligned SDF (HSDF) that effectively integrates geometric structure with lane semantic priors. By combining differentiable rendering to generate accurate height maps and designing a lane-aware semantic positional encoding (LSPE) to guide Transformer queries, the method eliminates reliance on conventional anchor-based regression and planar road assumptions. Evaluated on the OpenLane benchmark, the proposed framework achieves state-of-the-art performance in both 3D lane detection and height estimation.

Abstract

Monocular 3D lane detection plays a critical role in autonomous driving, yet recovering reliable 3D geometry from a single image remains challenging due to inherent depth ambiguity. Prior methods project image features into Bird's-Eye-View (BEV) space under a flat-ground assumption, causing geometric distortion on real-world roads. Recent methods instead predict explicit height maps to capture non-planar surfaces, but still rely on sparse anchor-based regression and exploit the recovered geometry merely for spatial transformation rather than semantic understanding. To overcome these limitations, we propose HSDF-Lane, which implicitly models the road surface as a Height-aligned Signed Distance Field (HSDF) over a densely sampled 3D feature volume. Through differentiable rendering, the HSDF jointly produces an accurate height map and surface-aligned features. We further introduce Lane-aware Semantic Positional Encoding (LSPE), which injects a lane-existence prior derived from the surface-aligned features into the transformer queries, coupling geometric structure with semantic guidance. Extensive experiments on the OpenLane benchmark show that HSDF-Lane achieves state-of-the-art performance in both 3D lane detection and height map estimation.

Problem

• 3D lane detection
• depth ambiguity
• non-planar road surfaces
• geometric distortion
• semantic understanding

Innovation

• Signed Distance Field
• 3D Lane Detection
• Height-Aligned Representation
• Semantic Positional Encoding
• Differentiable Rendering