Semantic segmentation of surface from lidar point cloud

Lecture Notes in Computer Science, 2019

Mapping the environment has been an important task for robot navigation and Simultaneous Localization And Mapping (SLAM). LIDAR provides a fast and accurate 3D point cloud map of the environment which helps in map building. However, processing millions of points in the point cloud becomes a computationally expensive task. In this paper, a methodology is presented to generate the segmented surfaces in real time and these can be used in modeling the 3D objects. At first an algorithm is proposed for efficient map building from single shot data of spinning Lidar. It is based on fast meshing and sub-sampling. It exploits the physical design and the working principle of the spinning Lidar sensor. The generated mesh surfaces are then segmented by estimating the normal and considering their homogeneity. The segmented surfaces can be used as proposals for predicting geometrically accurate model of objects in the robots activity environment. The proposed methodology is compared with some popular point cloud segmentation methods to highlight the efficacy in terms of accuracy and speed.

Applied Sciences, 2020

This paper presents a study of how the performance of LiDAR odometry is affected by the preprocessing of the point cloud through the use of 3D semantic segmentation. The study analyzed the estimated trajectories when the semantic information is exploited to filter the original raw data. Different filtering configurations were tested: raw (original point cloud), dynamic (dynamic obstacles are removed from the point cloud), dynamic vehicles (vehicles are removed), far (distant points are removed), ground (the points belonging to the ground are removed) and structure (only structures and objects are kept in the point cloud). The experiments were performed using the KITTI and SemanticKITTI datasets, which feature different scenarios that allowed identifying the implications and relevance of each element of the environment in LiDAR odometry algorithms. The conclusions obtained from this work are of special relevance for improving the efficiency of LiDAR odometry algorithms in all kinds o...

Geocarto International, 2018

Semantic labelling of LiDAR point cloud is critical for effective utilization of 3D points in numerous applications. 3D segmentation, incorporation of ancillary data, feature extraction and classification are the key stages in object-based point cloud labelling. The choice of algorithms and tuning parameters adopted in these stages has substantial impact on the quality of results from object-based point cloud labelling.This paper critically evaluates the performance of object-based point cloud labelling as a function of different 3D segmentation approaches, incorporation of spectral data, and computational complexity of the point cloud. The designed experiments are implemented on the datasets provided by the ISPRS and the results are independently validated by the ISPRS. Results indicate that aggregation of dense point cloud into higher-level object analogue (e.g. supervoxels) before 3D segmentation stage offers superior labelling results and best computational performance compared to the popular surface growing based approaches.

2019

This paper proposes RIU-Net (for Range-Image U-Net), the adaptation of a popular semantic segmentation network for the semantic segmentation of a 3D LiDAR point cloud. The point cloud is turned into a 2D range-image by exploiting the topology of the sensor. This image is then used as input to a U-net. This architecture has already proved its efficiency for the task of semantic segmentation of medical images. We demonstrate how it can also be used for the accurate semantic segmentation of a 3D LiDAR point cloud and how it represents a valid bridge between image processing and 3D point cloud processing. Our model is trained on range-images built from KITTI 3D object detection dataset. Experiments show that RIU-Net, despite being very simple, offers results that are comparable to the state-of-the-art of range-image based methods. Finally, we demonstrate that this architecture is able to operate at 90fps on a single GPU, which enables deployment for real-time segmentation.

IEEE Access

LiDARs are one of the key sources of reliable environmental ranging information for autonomous vehicles. However, segmentation of 3D scene elements (roads, buildings, people, cars, etc.) based on LiDAR point clouds has limitations. On the one hand, point-and voxel-based segmentation neural networks do not offer sufficiently high speed. On the other hand, modern labeled datasets primarily consist of street scenes recorded for driverless cars and contain little data for mobile delivery robots or cleaners that must work in parks and yards with heavy pedestrian traffic. This article aims to overcome these limitations. We have proposed a novel approach called DAPS3D to train deep neural networks for 3D semantic segmentation. This approach is based on a spherical projection of a point cloud and LiDAR-specific masks, enabling the model to adapt to different types of LiDAR. First of all, we have introduced various high-speed multi-scale spherical projection segmentation models, including convolutional, recurrent, and transformer architectures. Among them, the SalsaNextRecLSTM model has shown the best results. Secondly, we have proposed several original augmentations for spherical projections of LiDAR data, including FoV, flip, and rotation augmentation, as well as a special T-Zone cutout. These augmentations increase the model's invariance when dealing with changes in the data domain. Finally, we introduce a new method to generate synthetic datasets for domain adaptation problems. We have developed two new datasets for validating 3D scene outdoor segmentation algorithms: the DAPS-1 dataset, which is based on the augmentation of the reconstructed 3D semantic map, and the DAPS-2 LiDAR dataset, collected by the on-board sensors of a cleaning robot in a park area. Particular attention is given to the performance of the developed models, demonstrating their ability to function in real-time. The code and datasets used in this study are publicly available at: github.com/subake/DAPS3D.

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020

Black Sea Journal of Engineering and Science, 2020

In recent years, point cloud data generated with RGB-D cameras, 3D lasers, and 3D LiDARs have been employed frequently in robotic applications. In indoor environments, RGB-D cameras, which have short-range and can only describe the vicinity of the robots, generally are opted due to their low cost. On the other hand, 3D lasers and LiDARs can capture long-range measurements and generally are used in outdoor applications. In this study, we deal with the segmentation of indoor planar surfaces such as wall, floor, and ceiling via point cloud data. The segmentation methods, which are situated in Point Cloud Library (PCL) were executed with 3D laser point cloud data. The experiments were conducted to evaluate the performance of these methods with the publicly available Fukuoka indoor laser dataset, which has point clouds with different noise levels. The test results were compared in terms of segmentation accuracy and the time elapsed for segmentation. Besides, the general characteristics of each method were discussed. In this way, we revealed the positive and negative aspects of these methods for researchers that plan to apply them to 3D laser point cloud data.

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), 2019

We propose LU-Net-for LiDAR U-Net, a new method for the semantic segmentation of a 3D LiDAR point cloud. Instead of applying some global 3D segmentation method such as PointNet, we propose an end-to-end architecture for LiDAR point cloud semantic segmentation that efficiently solves the problem as an image processing problem. We first extract high-level 3D features for each point given its 3D neighbors. Then, these features are projected into a 2D multichannel range-image by considering the topology of the sensor. Thanks to these learned features and this projection, we can finally perform the segmentation using a simple U-Net segmentation network, which performs very well while being very efficient. In this way, we can exploit both the 3D nature of the data and the specificity of the Li-DAR sensor. This approach outperforms the state-of-the-art by a large margin on the KITTI dataset, as our experiments show. Moreover, this approach operates at 24fps on a single GPU. This is above the acquisition rate of common LiDAR sensors which makes it suitable for real-time applications.

IPCV'13 The 2013 International Conference on Image Processing, Computer Vision, and Pattern Recognition, 2013

We present algorithms for fast segmentation and classification of sparse 3D point clouds from rotating LIDAR sensors used for real-time applications such as autonomous mobile systems. Such systems must continuously process large amounts of data with update rates as high as 10 frames per second which makes complexity and performance of the algorithms very critical. Our approach to the segmentation of large and sparse point clouds is efficient and accurate which frees system resources for implementing other more demanding tasks such as classification. Segmentation is the emphasis of this paper as a necessary important first step for subsequent classification and further processing. We propose methods for segmenting sparse point clouds from rotating LIDAR sensors such as the Velodyne HDL-64E using rectangular and radial grids. The main part of the segmentation is performed on small grid images instead of large point clouds which makes the process computationally fast, simple, and very efficient. The proposed algorithms do not require flat horizontal structure of the ground and they demonstrate stable performance in different urban conditions and scenes for detection of different street objects including pedestrians, cars, and bicyclists.

IEEE Access

In the autonomous car, perception with point cloud semantic segmentation helps obtain a wealth of information about the surrounding road environment. Despite the massive progress of recent researches, the existing machine learning networks are still insufficient for online applications of autonomous driving due to too subdivided classes, the lack of training data, and their heavy computing load. To solve these problems, we propose a fast and lite point cloud semantic segmentation network for autonomous driving, which utilizes LiDAR synthetic data to improve the performance by transfer learning. First, we modify the labeling classes and generate the LiDAR synthetic data-set for additional training to alleviate the lack of training data of the realistic data-set. Then, to lower the computing load, we adopt a projection-based method and apply a lightweight segmentation network to projected LiDAR images, which has drastically reduced computing. Finally, we verified and evaluated the proposed network in this paper through experiments. Experimental results show that the proposed network can perform the three-dimensional point cloud semantic segmentation in an online way, in which the inference speed overwhelms the existing algorithms. INDEX TERMS Autonomous driving, point cloud semantic segmentation, synthetic dataset, computing load reduction.