Fast LIDAR-based road detection using fully convolutional neural networks
Paper in proceeding, 2017

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.

Semantics

Image coding

Intelligent vehicle highway systems

Transportation

Benchmarking

Neural networks

Convolution

Optical radar

Author

Luca Caltagirone

Chalmers, Applied Mechanics, Vehicle Engineering and Autonomous Systems

Samuel Scheidegger

Chalmers, Signals and Systems

Lennart Svensson

Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering

Mattias Wahde

Chalmers, Applied Mechanics, Vehicle Engineering and Autonomous Systems

IEEE Intelligent Vehicles Symposium, Proceedings

1019-1024

COPPLAR CampusShuttle cooperative perception & planning platform

VINNOVA (2015-04849), 2016-01-01 -- 2018-12-31.

Areas of Advance

Transport

Subject Categories

Computer Science

Computer Systems

Computer Vision and Robotics (Autonomous Systems)

DOI

10.1109/IVS.2017.7995848

More information

Latest update

5/10/2019