Convolutional Recurrent Network for Road Boundary Extraction

TL;DR

This paper presents a fully automatic convolutional recurrent network that accurately extracts road boundaries from LiDAR and camera data, significantly improving high definition map creation for autonomous driving.

Contribution

The novel structured model combines convolutional and recurrent networks to produce precise, topologically correct road boundary polylines without human intervention.

Findings

Achieves 99.3% perfect topology accuracy

High precision and recall in boundary extraction

Effective on large North American city dataset

Abstract

Creating high definition maps that contain precise information of static elements of the scene is of utmost importance for enabling self driving cars to drive safely. In this paper, we tackle the problem of drivable road boundary extraction from LiDAR and camera imagery. Towards this goal, we design a structured model where a fully convolutional network obtains deep features encoding the location and direction of road boundaries and then, a convolutional recurrent network outputs a polyline representation for each one of them. Importantly, our method is fully automatic and does not require a user in the loop. We showcase the effectiveness of our method on a large North American city where we obtain perfect topology of road boundaries 99.3% of the time at a high precision and recall.