LiDAR object detection algorithms based on neural networks for autonomous driving require large amounts of data for training, validation, and testing. As real-world data collection and labeling are time-consuming and expensive, simulation-based synthetic data generation is a viable alternative. However, using simulated data for the training of neural networks leads to a domain shift of training and testing data due to differences in scenes, scenarios, and distributions. In this work, we quantify the sim-to-real domain shift by means of LiDAR object detectors trained with a new scenario-identical real-world and simulated dataset. In addition, we answer the questions of how well the simulated data resembles the real-world data and how well object detectors trained on simulated data perform on real-world data. Further, we analyze point clouds at the target-level by comparing real-world and simulated point clouds within the 3D bounding boxes of the targets. Our experiments show that a significant sim-to-real domain shift exists even for our scenario-identical datasets. This domain shift amounts to an average precision reduction of around 14 % for object detectors trained with simulated data. Additional experiments reveal that this domain shift can be lowered by introducing a simple noise model in simulation. We further show that a simple downsampling method to model real-world physics does not influence the performance of the object detectors.
In this work, we shed light on different data augmentation techniques commonly used in Light Detection and Ranging (LiDAR) based 3D Object Detection. For the bulk of our experiments, we utilize the well known PointPillars pipeline and the well established KITTI dataset. We investigate a variety of global and local augmentation techniques, where global augmentation techniques are applied to the entire point cloud of a scene and local augmentation techniques are only applied to points belonging to individual objects in the scene. Our findings show that both types of data augmentation can lead to…
This paper presents a novel indoor layout estimation system based on the fusion of 2D LiDAR and intensity camera data. A ground robot explores an indoor space with a single floor and vertical walls, and collects a sequence of intensity images and 2D LiDAR datasets. The LiDAR provides accurate depth information, while the camera captures high-resolution data for semantic interpretation. The alignment of sensor outputs and image segmentation are computed jointly by aligning LiDAR points, as samples of the room contour, to ground-wall boundaries in the images. The alignment problem is decoupled i…
Seoul National University（首尔国立大学）；Agency for Defense Development（国防发展局）；Inha University（仁荷大学）
LiDAR is widely used to capture accurate 3D outdoor scene structures. However, LiDAR produces many undesirable noise points in snowy weather, which hamper analyzing meaningful 3D scene structures. Semantic segmentation with snow labels would be a straightforward solution for removing them, but it requires laborious point-wise annotation. To address this problem, we propose a novel self-supervised learning framework for snow points removal in LiDAR point clouds. Our method exploits the structural characteristic of the noise points: low spatial correlation with their neighbors. Our method consis…
A critical aspect of autonomous vehicles (AVs) is the object detection stage, which is increasingly being performed with sensor fusion models: multimodal 3D object detection models which utilize both 2D RGB image data and 3D data from a LIDAR sensor as inputs. In this work, we perform the first study to analyze the robustness of a high-performance, open source sensor fusion model architecture towards adversarial attacks and challenge the popular belief that the use of additional sensors automatically mitigate the risk of adversarial attacks. We find that despite the use of a LIDAR sensor, the …
Daimler AG（戴姆勒股份公司）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；FZI Research Center for Information Technology（FZI信息技术研究中心）
State-of-the-art approaches for the semantic labeling of LiDAR point clouds heavily rely on the use of deep Convolutional Neural Networks (CNNs). However, transferring network architectures across different LiDAR sensor types represents a significant challenge, especially due to sensor specific design choices with regard to network architecture as well as data representation. In this paper we propose a new CNN architecture for the point-wise semantic labeling of LiDAR data which achieves state-of-the-art results while increasing portability across sensor types. This represents a significant ad…
Modern perception systems in the field of autonomous driving rely on 3D data analysis. LiDAR sensors are frequently used to acquire such data due to their increased resilience to different lighting conditions. Although rotating LiDAR scanners produce ring-shaped patterns in space, most networks analyze their data using an orthogonal voxel sampling strategy. This work presents a novel approach for analyzing 3D data produced by 360-degree depth scanners, utilizing a more suitable coordinate system, which is aligned with the scanning pattern. Furthermore, we introduce a novel notion of range-guid…
Analyzing Infrastructure LiDAR Placement with Realistic LiDAR Simulation Library
Shanghai AI Laboratory（上海人工智能实验室）；Institute of Artificial Intelligence, Beihang University（北京航空航天大学人工智能研究院）；University of California, Los Angeles（加利福尼亚大学洛杉矶分校）
Over the last decade, robotic perception algorithms have significantly benefited from the rapid advances in deep learning (DL). Indeed, a significant amount of the autonomy stack of different commercial and research platforms relies on DL for situational awareness, especially vision sensors. This work explores the potential of general-purpose DL perception algorithms, specifically detection and segmentation neural networks, for processing image-like outputs of advanced lidar sensors. Rather than processing the three-dimensional point cloud data, this is, to the best of our knowledge, the first…
Karlsruhe Institute of Technology (KIT)（卡尔斯鲁厄理工学院）；FZI Research Center for Information Technology（FZI信息技术研究中心）
We present a generic evidential grid mapping pipeline designed for imaging sensors such as LiDARs and cameras. Our grid-based evidential model contains semantic estimates for cell occupancy and ground separately. We specify the estimation steps for input data represented by point sets, but mainly focus on input data represented by images such as disparity maps or LiDAR range images. Instead of relying on an external ground segmentation only, we deduce occupancy evidence by analyzing the surface orientation around measurements. We conduct experiments and evaluate the presented method using LiDA…
Visual data in autonomous driving perception, such as camera image and LiDAR point cloud, can be interpreted as a mixture of two aspects: semantic feature and geometric structure. Semantics come from the appearance and context of objects to the sensor, while geometric structure is the actual 3D shape of point clouds. Most detectors on LiDAR point clouds focus only on analyzing the geometric structure of objects in real 3D space. Unlike previous works, we propose to learn both semantic feature and geometric structure via a unified multi-view framework. Our method exploits the nature of LiDAR sc…
Digitisation of fruit trees using LiDAR enables analysis which can be used to better growing practices to improve yield. Sophisticated analysis requires geometric and semantic understanding of the data, including the ability to discern individual trees as well as identifying leafy and structural matter. Extraction of this information should be rapid, as should data capture, so that entire orchards can be processed, but existing methods for classification and segmentation rely on high-quality data or additional data sources like cameras. We present a method for analysis of LiDAR data specifical…
Mercedes-Benz AG（梅赛德斯-奔驰股份公司）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；Research Center for Information Technology（信息技术研究中心）
Judging the quality of samples synthesized by generative models can be tedious and time consuming, especially for complex data structures, such as point clouds. This paper presents a novel approach to quantify the realism of local regions in LiDAR point clouds. Relevant features are learned from real-world and synthetic point clouds by training on a proxy classification task. Inspired by fair networks, we use an adversarial technique to discourage the encoding of dataset-specific information. The resulting metric can assign a quality score to samples without requiring any task specific annotat…
Technical Report: Quantifying and Analyzing the Generalization Power of a DNN
University of Electronic Science and Technology of China（电子科技大学）；Shanghai Jiao Tong University（上海交通大学）；Institute of Artificial Intelligence, China Telecom（中国电信人工智能研究院）
The ability to simultaneously leverage multiple modes of sensor information is critical for perception of an automated vehicle's physical surroundings. Spatio-temporal alignment of registration of the incoming information is often a prerequisite to analyzing the fused data. The persistence and reliability of multi-modal registration is therefore the key to the stability of decision support systems ingesting the fused information. LiDAR-video systems like on those many driverless cars are a common example of where keeping the LiDAR and video channels registered to common physical features is im…
Shanghai Jiao Tong University（上海交通大学）；Huazhong University of Science & Technology（华中科技大学）；Xi’an Jiaotong University（西安交通大学）
This paper presents a method to interpret the success of knowledge distillation by quantifying and analyzing task-relevant and task-irrelevant visual concepts that are encoded in intermediate layers of a deep neural network (DNN). More specifically, three hypotheses are proposed as follows. 1. Knowledge distillation makes the DNN learn more visual concepts than learning from raw data. 2. Knowledge distillation ensures that the DNN is prone to learning various visual concepts simultaneously. Whereas, in the scenario of learning from raw data, the DNN learns visual concepts sequentially. 3. Know…
Airborne LiDAR point cloud representing a forest contains 3D data, from which vertical stand structure even of understory layers can be derived. This paper presents a tree segmentation approach for multi-story stands that stratifies the point cloud to canopy layers and segments individual tree crowns within each layer using a digital surface model based tree segmentation method. The novelty of the approach is the stratification procedure that separates the point cloud to an overstory and multiple understory tree canopy layers by analyzing vertical distributions of LiDAR points within overlappi…
Neural varifolds: an aggregate representation for quantifying the geometry of point clouds
School of Electronics and Computer Science, University of Southampton（南安普顿大学电子与计算机科学学院）；Department of Applied Mathematics and Theoretical Physics, University of Cambridge（剑桥大学应用数学与理论物理系）；Institut Camille Jordan, Université Claude Bernard Lyon 1（里昂第一大学卡米尔·约旦研究所）
Methods for the Segmentation of Reticular Structures Using 3D LiDAR Data: A Comparative Evaluation
Engineering Research Institute of Elche (I3E)（埃尔切工程研究所 (I3E)）；Valencian Graduate School and Research Network of Artificial Intelligence (ValgrAI)（瓦伦西亚人工智能研究生院与研究网络 (ValgrAI)）
Recent advances in digital imaging, e.g., increased number of pixels captured, have meant that the volume of data to be processed and analyzed from these images has also increased. Deep learning algorithms are state-of-the-art for analyzing such images, given their high accuracy when trained with a large data volume of data. Nevertheless, such analysis requires considerable computational power, making such algorithms time- and resource-demanding. Such high demands can be met by using third-party cloud service providers. However, analyzing medical images using such services raises several legal…
