Search

LiDAR is one of the main sensors for 3D object detection in autonomous driving. LiDAR has the advantages of high precision and high resolution, but as the distance increases, the points it acquires become sparse, resulting in uneven sampling points and hindering the feature extraction of discrete objects. Current 3D object detection methods using LiDAR ignore the sparse features of the original LiDAR point cloud, resulting in low classification accuracy over small object detection, hindering the development of autonomous driving technology. To address this problem, we propose point cloud Sparse Detection Network (PCSD), an end-to-end two-stage 3D object detection framework. First, PCSD uses a data augmentation algorithm to preprocess the KITTI dataset, then uses voxel point centroids to locate voxel features, and then uses a point sparsity-aware RoI grid pooling module to aggregate local voxel features. Finally, we improve the confidence of the final bounding box by using voxel features with the original point cloud sparse features. Experimental evaluation on the challenging KITTI object detection benchmark shows significant improvements, especially in pedestrian and cyclist classification accuracy improved by 13.22% and 9.33%, respectively, demonstrating the feasibility and applicability of our work. [ABSTRACT FROM AUTHOR]

A large amount of environmental point cloud data can be provided by light detection and ranging (LiDAR) sensors. The raw points collected by the LiDAR is disordered and unstructured. It is a challenge to design algorithms to extract features from the raw points. In this paper, we propose the LiDAR point semantic segmentation net (LPSS Net), which is a dual-attention mechanism point cloud segmentation algorithm. First, the LPSS Net extracts point cloud features from the raw points, which uses the self-attention mechanism in the transformer mechanism. Second, in order to suppress irrelevant information in the features and focus on essential information, we propose a novel strategy, which designs a 3D channel attention mechanism in the encoding part. Finally, we demonstrate the applicability of the algorithm by extensive experiments conducted on Semantic3D and SemanticKITTI. A high OA of 90.6% and a mIoU of 71.5% on the Semantic3D data set indicate the feasibility of the algorithm. [ABSTRACT FROM AUTHOR]

Light detection and ranging (LiDAR) camera systems are becoming increasingly vital for autonomous driving. The monocular three-dimensional (3D) detection task is a critical and challenging aspect of this field. However, most algorithms rely solely on manually labeled images, which is a time-consuming and labor-intensive process, and the resulting detection lacks depth information. To address this problem, a semi-supervised 3D object detection model based on LiDAR camera systems (PVONet) is proposed to improve both the detection accuracy and processing time. First, an innovative data preparation block point-voxel fusion estimation is introduced; it utilizes LiDAR points to generate 3D bounding boxes for unlabeled data, thereby significantly reducing the time compared with manual labeling. Second, a new block based on fully connected neural network for box estimation (feature extraction and 3D object detection) is presented; it is utilized to conduct feature extraction, feature correlation, and 3D box estimation on monocular images. Finally, comprehensive experiments conducted on the popular KITTI 3D detection dataset demonstrate that our PVONet is faster (30 ms on KITTI benchmark) and more accurate [with increases of 4.69%/3.82% (easy), 4.45%/2.79% (moderate), and 4.07%/3.75% (hard) aggregation processes on 3D/bird's eye view objects compared with the baseline]. This meets the requirements for high real-time performance in autonomous vehicles applications. The results demonstrate the effectiveness of our model based on LiDAR camera systems. [ABSTRACT FROM AUTHOR]