CSPC-Dataset: New LiDAR Point Cloud Dataset and Benchmark for Large-Scale Scene Semantic Segmentation

Abstract

Large-scale point clouds scanned by light detection and ranging (lidar) sensors provide detailed geometric characteristics of scenes due to the provision of 3D structural data. The semantic segmentation of large-scale point clouds is a crucial step for an in-depth understanding of complex scenes. Of late, although a large number of point cloud semantic segmentation algorithms have been proposed, semantic segmentation methods are still far from being satisfactory in terms of precision and accuracy of large-scale point clouds. For machine learning (ML) and deep learning (DL) methodologies, the semantic segmentation is largely influenced by the quality of training sets and methods themselves. Therefore, we construct a new point cloud dataset, namely CSPC-Dataset (Complex Scene Point Cloud Dataset) for large-scale scene semantic segmentation. CSPC-Dataset point clouds are acquired by a wearable laser mobile mapping robot. It covers five complex urban and rural scenes and mainly includes six types of objects, i.e., ground, car, building, vegetation, bridge, and pole. It provides large-scale outdoor scenes with color information, which has advantages such as the scene more complete, point density relatively uniform, diversity and complexity of objects and the high discrepancy between different scenes. Based on the CSPC-Dataset, we construct a new benchmark, which includes approximately 68 million points with explicit semantic labels. To extend the dataset into a wide range of applications, this paper provides the semantic segmentation results and comparative analysis of 7 baseline methods based on CSPC-Dataset. In the experiment part, three groups of experiments are conducted for benchmarking, which offers an effective way to make comparisons with different point-labeling algorithms. The labeling results have shown that the highest Intersection over Union (IoU) of pole, ground, building, car, vegetation, and bridge for all benchmarks is 36.0%, 97.8%,

Multi-Scale Attentive Aggregation for LiDAR Point Cloud Segmentation

“…We evaluated the proposed method on two datasets, CSPC [38] and Toronto3D [39]. CSPC (Complex Scene Point Cloud dataset) is the most recent point cloud dataset for semantic segmentation of large-scale outdoor scenes, covering five urban and rural scenes where scene-1 shows a simple street, scene-2 shows a busy urban street, scene-3 shows a busy urban street at night, scene-4 shows a campus, and scene-5 shows a rural street.…”

Point Cloud Classification Algorithm Based on the Fusion of the Local Binary Pattern Features and Structural Features of Voxels

“…Collection equipment of a point cloud scene is shown in Figure 3. Scene 1, Scene 2, and Scene 3 are outdoor scene colored point cloud data collected by advanced backpack mobile surveying and mapping robots provided in CSPC-Dataset [23]. The robot collects the data of these scenes by laser sensors and panoramic cameras, and the average point cloud density is about 50~60/m 2 .…”

The role of machine intelligence in photogrammetric 3D modeling – an overview and perspectives

“…Although it is no longer the top-performing approach, it drives many other methods that bring the capacity of approaches to a notable level. Today, the benchmark datasets and open competitions (such as IEEE data fusion contest (Le Saux et al, 2019)) explicitly designed tasks with training data that match the needed training data volume for deep learning models (Hackel et al, 2017;Niemeyer et al, 2014;Tong et al, 2020), in which the traditional methods (shallow classifiers) are somewhat less competent. Nevertheless, it should be noted that in practical applications, the volume of available data for training is still of critical concern, which might be insufficient to drive deep learning models.…”