Anchor3DLane++: 3D Lane Detection via Sample-Adaptive Sparse 3D Anchor Regression

TL;DR

Anchor3DLane++ introduces a BEV-free 3D lane detection method that uses sample-adaptive sparse anchors and fusion techniques, outperforming previous approaches on standard benchmarks.

Contribution

The paper proposes a novel BEV-free 3D lane detection framework with dynamic anchor generation and regularization, advancing beyond prior BEV-based and direct FV methods.

Findings

Outperforms state-of-the-art methods on three benchmarks.

Introduces a dynamic anchor generation module (PAAG).

Leverages camera-LiDAR fusion for improved accuracy.

Abstract

In this paper, we focus on the challenging task of monocular 3D lane detection. Previous methods typically adopt inverse perspective mapping (IPM) to transform the Front-Viewed (FV) images or features into the Bird-Eye-Viewed (BEV) space for lane detection. However, IPM's dependence on flat ground assumption and context information loss in BEV representations lead to inaccurate 3D information estimation. Though efforts have been made to bypass BEV and directly predict 3D lanes from FV representations, their performances still fall behind BEV-based methods due to a lack of structured modeling of 3D lanes. In this paper, we propose a novel BEV-free method named Anchor3DLane++ which defines 3D lane anchors as structural representations and makes predictions directly from FV features. We also design a Prototype-based Adaptive Anchor Generation (PAAG) module to generate sample-adaptive sparse 3D anchors dynamically. In addition, an Equal-Width (EW) loss is developed to leverage the parallel property of lanes for regularization. Furthermore, camera-LiDAR fusion is also explored based on Anchor3DLane++ to leverage complementary information. Extensive experiments on three popular 3D lane detection benchmarks show that our Anchor3DLane++ outperforms previous state-of-the-art methods. Code is available at: https://github.com/tusen-ai/Anchor3DLane.