Fast LIDAR-based Road Detection Using Fully Convolutional Neural Networks

TL;DR

This paper presents a fast, LIDAR-based road detection method using a fully convolutional neural network that transforms point clouds into top-view images, enabling real-time semantic segmentation for autonomous driving.

Contribution

The work introduces a novel top-view representation from LIDAR data and a specialized FCN architecture for efficient, accurate road detection suitable for real-time use.

Findings

Achieved top performance on the KITTI benchmark

Demonstrated real-time inference capability

Outperformed existing LIDAR-based road detection methods

Abstract

In this work, a deep learning approach has been developed to carry out road detection using only LIDAR data. Starting from an unstructured point cloud, top-view images encoding several basic statistics such as mean elevation and density are generated. By considering a top-view representation, road detection is reduced to a single-scale problem that can be addressed with a simple and fast fully convolutional neural network (FCN). The FCN is specifically designed for the task of pixel-wise semantic segmentation by combining a large receptive field with high-resolution feature maps. The proposed system achieved excellent performance and it is among the top-performing algorithms on the KITTI road benchmark. Its fast inference makes it particularly suitable for real-time applications.