3D-LaneNet+: Anchor Free Lane Detection using a Semi-Local Representation

TL;DR

3D-LaneNet+ is an advanced anchor-free deep learning method for 3D lane detection that effectively handles complex lane topologies, improving generalization and accuracy over previous models.

Contribution

It introduces a semi-local tile representation and feature embedding to detect complex lane structures without anchors or lane fitting.

Findings

Significant performance improvement over 3D-LaneNet.

Effective detection of arbitrary lane topologies.

Better generalization to complex geometries.

Abstract

3D-LaneNet+ is a camera-based DNN method for anchor free 3D lane detection which is able to detect 3d lanes of any arbitrary topology such as splits, merges, as well as short and perpendicular lanes. We follow recently proposed 3D-LaneNet, and extend it to enable the detection of these previously unsupported lane topologies. Our output representation is an anchor free, semi-local tile representation that breaks down lanes into simple lane segments whose parameters can be learnt. In addition we learn, per lane instance, feature embedding that reasons for the global connectivity of locally detected segments to form full 3d lanes. This combination allows 3D-LaneNet+ to avoid using lane anchors, non-maximum suppression, and lane model fitting as in the original 3D-LaneNet. We demonstrate the efficacy of 3D-LaneNet+ using both synthetic and real world data. Results show significant improvement relative to the original 3D-LaneNet that can be attributed to better generalization to complex lane topologies, curvatures and surface geometries.