Polar R-CNN: End-to-End Lane Detection with Fewer Anchors

TL;DR

Polar R-CNN introduces an end-to-end lane detection approach that reduces reliance on dense anchors and NMS, using polar coordinate systems to improve efficiency and performance in complex driving scenarios.

Contribution

The paper presents Polar R-CNN, a novel lane detection method that minimizes anchor usage and eliminates NMS, enhancing deployment efficiency and adaptability across diverse datasets.

Findings

Achieves competitive results on five lane detection benchmarks.

Reduces number of anchors needed without performance loss.

Supports NMS-free detection for better deployment efficiency.

Abstract

Lane detection is a critical and challenging task in autonomous driving, particularly in real-world scenarios where traffic lanes can be slender, lengthy, and often obscured by other vehicles, complicating detection efforts. Existing anchor-based methods typically rely on prior lane anchors to extract features and subsequently refine the location and shape of lanes. While these methods achieve high performance, manually setting prior anchors is cumbersome, and ensuring sufficient coverage across diverse datasets often requires a large amount of dense anchors. Furthermore, the use of Non-Maximum Suppression (NMS) to eliminate redundant predictions complicates real-world deployment and may underperform in complex scenarios. In this paper, we propose Polar R-CNN, an end-to-end anchor-based method for lane detection. By incorporating both local and global polar coordinate systems, Polar R-CNN facilitates flexible anchor proposals and significantly reduces the number of anchors required without compromising performance.Additionally, we introduce a triplet head with heuristic structure that supports NMS-free paradigm, enhancing deployment efficiency and performance in scenarios with dense lanes.Our method achieves competitive results on five popular lane detection benchmarks--Tusimple, CULane,LLAMAS, CurveLanes, and DL-Rai--while maintaining a lightweight design and straightforward structure. Our source code is available at https://github.com/ShqWW/PolarRCNN.