CPL - Chalmers Publication Library
| Utbildning | Forskning | Styrkeområden | Om Chalmers | In English In English Ej inloggad.

Fast LIDAR-based road detection using fully convolutional neural networks

Luca Caltagirone (Institutionen för tillämpad mekanik, Fordonsteknik och autonoma system) ; Samuel Scheidegger (Institutionen för signaler och system) ; Lennart Svensson (Institutionen för signaler och system, Signalbehandling) ; Mattias Wahde (Institutionen för tillämpad mekanik, Fordonsteknik och autonoma system)
IEEE Intelligent Vehicles Symposium, Proceedings p. 1019-1024. (2017)
[Konferensbidrag, refereegranskat]

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.

Nyckelord: Benchmarking; Convolution; Image coding; Intelligent vehicle highway systems; Neural networks; Optical radar; Semantics; Transportation

Denna post skapades 2017-10-04. Senast ändrad 2017-11-29.
CPL Pubid: 252307


Läs direkt!

Länk till annan sajt (kan kräva inloggning)

Institutioner (Chalmers)

Institutionen för tillämpad mekanik, Fordonsteknik och autonoma system (2010-2017)
Institutionen för signaler och system (1900-2017)
Institutionen för signaler och system, Signalbehandling (1900-2017)


Datavetenskap (datalogi)
Datorseende och robotik (autonoma system)

Chalmers infrastruktur