Semantic segmentation of outdoor scenes using LIDAR cloud point

Abstract

In this paper we present a novel street scene semantic recognition framework, which takes advantage of 3D point clouds captured by a high definition LiDAR laser scanner. An important problem in object recognition is the need for sufficient labeled training data to learn robust classifiers. In this paper we show how to significantly re-duce the need for manually labeled training data by reduction of scene complexity using non-supervised ground and building segmentation. Our system first automatically seg-ments grounds point cloud, this is because the ground connects almost all other objects and we will use a connect component based algorithm to over segment the point clouds. Then, using binary range image processing building facades will be detected. Remained point cloud will grouped into voxels which are then transformed to super voxels. Local 3D features extracted from super voxels are classified by trained boosted decision trees and labeled with semantic classes e.g. tree, pedestrian, car. Given labeled 3D points cloud and 2D image with known viewing camera pose, the proposed association module aligned collections of 3D points to the groups of 2D image pixel to parsing 2D cubic images. One noticeable advantage of our method is the robustness to different lighting condition, shadows and city landscape. The proposed method is evaluated both quantitatively and qualitatively on a challenging fixed-position Terrestrial Laser Scanning (TLS) Velodyne data set and Mobile Laser Scanning (MLS), NAVTEQ True databases. Robust scene parsing results are reported.