Lane detection is extremely important for autonomous vehicles. For this reason, many approaches use lane boundary information to locate the vehicle inside the street, or to integrate GPS-based localization. As many other computer vision based tasks, convolutional neural networks (CNNs) represent the state-of-the-art technology to indentify lane boundaries. However, the position of the lane boundaries w.r.t. the vehicle may not suffice for a reliable positioning, as for path planning or localization information regarding lane types may also be needed. In this work, we present an end-to-end system for lane boundary identification, clustering and classification, based on two cascaded neural networks, that runs in real-time. To build the system, 14336 lane boundaries instances of the TuSimple dataset for lane detection have been labelled using 8 different classes. Our dataset and the code for inference are available online.

No abstract

No abstract

Autonomous navigation relies upon an accurate understanding of the elements in the surroundings. Among the different on-board perception tasks, 3D object detection allows the identification of dynamic objects that cannot be registered by maps, being key for safe navigation. Thus, it often requires the use of LiDAR data, which is able to faithfully represent the scene geometry. However, although raw laser point clouds contain rich features to perform object detection, more compact representations such as the bird's eye view (BEV) projection are usually preferred in order to meet the time requirements of the control loop. This paper presents an end-to-end object detection network based on the well-known Faster R-CNN architecture that uses BEV images as input to produce the final 3D boxes. Our regression branches can infer not only the axis-aligned bounding boxes but also the rotation angle, height, and elevation of the objects in the scene. The proposed network provides state-of-the-art results for car, pedestrian, and cyclist detection with a single forward pass when evaluated on the KITTI 3D Object Detection Benchmark, with an accuracy that exceeds 64% mAP 3D for the Moderate difficulty. Further experiments on the challenging nuScenes dataset show the generalizability of both the method and the proposed BEV representation against different LiDAR devices and across a wider set of object categories by being able to reach more than 30% mAP with a single LiDAR sweep and almost 40% mAP with the usual 10-sweep accumulation.INDEX TERMS Bird's eye view (BEV), LiDAR, object detection, autonomous driving

LiDAR devices have become a key sensor for autonomous vehicles perception due to their ability to capture reliable geometry information. Indeed, approaches processing LiDAR data have shown an impressive accuracy for 3D object detection tasks, outperforming methods solely based on image inputs. However, the wide diversity of on-board sensor configurations makes the deployment of published algorithms into real platforms a hard task, due to the scarcity of annotated datasets containing laser scans. We present a method to generate new point clouds datasets as captured by a real LiDAR device. The proposed pipeline makes use of multiple frames to perform an accurate 3D reconstruction of the scene in the spherical coordinates system that enables the simulation of the sweeps of a virtual LiDAR sensor, configurable both in location and inner specifications. The similarity between real data and the generated synthetic clouds is assessed through a set of experiments performed using KITTI Depth and Object Benchmarks.