Fast LIDAR-based Road Detection Using Convolutional Neural Networks

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 height 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 convolutional neural network (CNN). The CNN 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 achieves state-of-the-art results on the KITTI road benchmark. It is currently the top-performing algorithm among the published methods in the overall category urban road and outperforms the second best LIDAR-only approach by 7.4 percentage points. Its fast inference makes it particularly suitable for real-time applications.